Professional Documents
Culture Documents
Thesis Wu en SCH 2008
Thesis Wu en SCH 2008
Signals in Networks
and
Related Graph Theoretical Problems on Spanning
Trees
vorgelegt von
Dipl.-Math. Gregor W
unsch
Vorsitzender:
Berichter:
Berlin 2008
D 83
ACKNOWLEDGEMENTS
This work would not have been possible without the help of a number of people. I
wish to thank everybody who has been involved, directly or indirectly.
First of all, I thank my supervisor Rolf Mohring. I am grateful for his support and
encouragement and I am especially indebted to him raising my interest for discrete
applied mathematics during my years as an undergraduate.
In addition, I wish to thank Ekkehard Kohler for taking the second assessment
of this thesis and for numerous discussions and stimulating ideas on traffic models
involving traffic signals.
Next, I wish to thank Klaus N
okel from PTV AG in Karlsruhe. During the last
approximately four years he and the PTV supported the work on an optimization
tool for coordinating signals in various ways. Not only that we were equipped with
software, but also I enjoyed our discussions and profited from valuable suggestions.
Also, I am very grateful for the financial support that I received within the
DFG research training group MAGSI (GK-621). I thank all my colleagues from
MAGSI for a fruitful interdisciplinary research environment. Moreover, I thank
Armin Zimmermann for organizing almost everything concerning the activities within
MAGSI.
Furthermore, I am indebted to my coauthors Ekki Kohler, Rolf Mohring, Klaus
Nokel, Alexander Reich and Romeo Rizzi and, especially, Christian Liebchen, who
raised my interest for strictly fundamental cycle bases. Yet, from all of them, I
learned a lot during intensive discussions.
I also want to thank all members of the research groups of Rolf Mohring, G
unter
Ziegler and Stefan Felsner for the excellent working atmosphere. Here, special thanks
go to my office mates Felix K
onig, Heiko Schilling and Bjorn Stenzel. I very much
enjoyed our numerous chats on various things like sports, elections, statistics and
discrete mathematics.
Moreover, I would like to thank Christian Liebchen, Nina Brenner, Tobias Harks,
Richard L
utjens, Nicole Megow, Guido Schafer, Izaskun Seara Tejados, Bjorn Stenzel, and Sebastian Stiller for their careful proofreading of the manuscript. The exposition of this thesis improved from their valuable suggestions and comments.
Last, but surely not least, I am grateful to my friends and my family for their
support and their interest in my work.
Berlin, February 2008
Gregor W
unsch
iii
CONTENTS
Introduction
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
Problem
. . . . .
. . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
7
7
7
9
11
13
13
15
17
19
20
21
24
27
28
31
36
37
39
.
.
.
.
.
.
.
.
41
41
44
45
45
57
60
66
70
vi
CONTENTS
2.4
2.5
2.6
Upper Bounds . . . . . . . . . . .
2.4.1 A New Asymptotical Upper
Experiments . . . . . . . . . . . . .
Conclusions and Open Questions .
. . . .
Bound
. . . .
. . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
Never Coincide
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
4 Experiments
4.1 A MIP Solver Comparison on Selected NSC Instances
4.2 The Influence of Cycle Bases on the MIP Performance
4.3 Case Studies . . . . . . . . . . . . . . . . . . . . . . .
4.3.1 Evaluating an NSC model . . . . . . . . . . . .
4.3.2 Data Acquisition . . . . . . . . . . . . . . . . .
4.3.3 Portland . . . . . . . . . . . . . . . . . . . . . .
4.3.4 Denver . . . . . . . . . . . . . . . . . . . . . . .
4.4 Conclusion and Open Questions . . . . . . . . . . . . .
Bibliography
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
77
79
84
87
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
91
91
93
96
96
99
101
101
102
108
110
110
114
114
.
.
.
.
.
.
.
.
117
. 117
. 123
. 128
. 128
. 130
. 132
. 136
. 139
143
INTRODUCTION
Are you the type of person that likes waiting at a red light? Do you enjoy being
stopped by uncoordinated signals? If not, you are invited to read on to find out how
the coordination of traffic signals can help to reduce delays and, thus, avoid having
to wait at red lights.
In urban areas there is a strong demand for transportation. Probably, the most
sustainable means of transportation are the public ones, like buses, trams or the
underground. Nevertheless, not all demands for transportation can be covered by
the public sector. A major part of the overall transportation in cities is composed of
individual drivers.
There are different ways to improve road traffic conditions in city areas. However,
infrastructural arrangements like broadening streets or even building new ones to
cope with increasing demands are often not an appropriate option, due to high costs
or space limitations. Instead, intelligent means of traffic control are required to solve
todays road traffic problems, like high delays, narrow capacities or traffic jams.
When speaking of ways to control traffic, traffic lights or traffic signals are of
primary importance. A clever adjustment of the signal settings surely helps to reduce delays and increase capacities, thereby avoiding traffic jams. Today, intelligent
computer-aided traffic control signals are even capable of reacting to different traffic situations. Namely, they adapt their settings to the respective demands at the
junctions.
However, there are traffic scenarios where the traffic responsive signals reach
their limit. For example, when there is constant and high traffic volume, the responsive signals repeatedly apply similar control strategies. Therefore, they behave
comparably to fixed time traffic signals. These fixed time traffic signals repeatedly
respond to a prescribed signal timing program and not to the actual traffic conditions. Hence, research into fixed time traffic signals and their control strategies is an
ongoing endeavor.
Operating fixed time signals offers different means of controlling traffic. On the
one hand, some signal parameters adjustments influence the traffic flow locally at a
single junction. In many situations, such a local calibration of the signals turns out
to be sufficient to cope with the aforementioned problems. On the other hand, in
1
Introduction
situations with constant high traffic, another non-local control strategy for the fixed
time signals becomes more significant: the coordination of the signals.
Coordinating traffic signals means the following: coupling of signals via a parameter called offset. This quantity specifies how green phases of different signals are
shifted (or offset) to each other. Most prominent coordination objectives are so-called
green waves, where vehicles travel without being impeded by a signal showing red.
Nevertheless, when considering networks of signals instead of arterials of signals, it
is often not possible to adjust green waves for the whole network. Instead, the goal
green wave has to be replaced by a more practical term like minimum possible
delay. Hereby, the item delay refers to waiting times of vehicles facing red at the
signals.
Many approaches and models have been proposed in order to find good coordinations of signals in networks. Still, the majority of them reveal shortcomings either
way, be it unrealistic modeling of real-world circumstances or the fact that they do
not give guarantees for their solution quality.
To summarize, there is a need for a mathematical optimization approach for
coordinating fixed time traffic signals in networks.
This discussion on new required control strategies for fixed time signals is not
a theoretical one. Rather, the industry, i.e., traffic companies that plan, manage,
and control traffic, demands applicable approaches for coordinating traffic signals in
networks.
As an indication thereof, we briefly report on an industry project that emerged between the TU Berlin and the PTV AG, which is a traffic planning software company
from Karlsruhe, Germany. In this project, the aim was to develop mathematical
optimization software to coordinate fixed time traffic signals in networks. During
the project, we developed a mixed-integer linear programming approach, which minimizes the delay of vehicles in a network by adjusting optimal offsets. However,
several other functionalities were incorporated in the model. The outcome of the
project with the PTV, though, is a concrete implementation of the optimization
approach, which is about to be included in PTV software soon.
In our mixed-integer linear program (MIP) for the coordination of traffic signals,
a particular physical constraint has been modeled. This constraint, which we will
therefore call Cycle Constraint, has to be formulated for all cycles C of the
graph G that represents the network of signalized junctions. It suffices, however, to
state the cycle constraints for the elements of a cycle basis. This then implies these
constraints for all cycles of G. Depending on the respective application, though, it
has to be a cycle basis with a certain property. In our case of a MIP for coordinating
fixed time signals in networks one has to define the cycle constraints for the elements
of an integral cycle basis.
This means that any integral cycle basis can be used to define the cycle constraints
for our MIP. Although any two integral cycle bases lead to MIPs with equal optimal
objective value, the computational behavior of their MIPs may be different. Observe
that this may be of importance, since we are considering networks of large size where
one may not come up with optimal solutions.
Introduction
A quantity one can use to compare cycle bases is the so-called width of a basis.
Loosely speaking, the width of a basis is defined as the productover all cycles
of a basisof the number of possible values that the integer variable for the cycle
constraint for that cycle can take. Thus, the width of a basis gives an impression of
the size of the MIPs feasibility region. The hope is that the smaller the width of a
basis, the better the corresponding MIP performs.
Among the class of integral cycle bases strictly fundamental cycle bases are a
prominent subclass. For a graph G = (V, E), a strictly fundamental cycle basis B is
defined by a spanning tree T of G. In particular, the cycles of B are exactly the ones
induced by non-tree edges of T with respect to the graph G. Then, in the Minimum
Strictly Fundamental Cycle Bases (MSFCB) problem one seeks a spanning tree T
that induces a basis of minimum length.
In 1982, Deo et al. [DKP82] proved the MSFCB problem to be NP-complete
for general graphs. Since then, many heuristics for the MSFCB problem have been
proposed. Nevertheless, for comparing the results of these heuristics, i.e., whenever
concrete experiments were conducted, sample graph classes were considered. Besides
random graphs, grid graphs are the most important such graph class.
Grid graphs are also of interest for the two following reasons. First, considering
the coordination of traffic signals, many real-world networks have a grid-like structure. One only has to think of the layout of central areas in north american cities.
Second, for the MSFCB problem, grid graphs turn out to be computationally tricky.
This fact is probably due to an extreme amount of symmetric spanning trees on
grids.
In 1995, Alon et al. [AKPW95] proved that for square grids with n vertices, the
size of an optimal solution to the MSFCB problem is in (n log n). Still, we decided
to investigate bounds on the optimal value of an MSFCB on a square grid having
the form
c1 n log2 n o(n log n) OPTn c2 n log2 n + o(n log n).
We could prove that the above statement is true for c1 = 1/12 and c2 = 0.979,
respectively.
An optimization problem that is closely related to the MSFCB problem is the
one of finding a t-tree spanner with minimal t, [CC95]. In this problem, one seeks a
spanning tree T for a given general graph G, such that the maximum over all pairs
of vertices (u, v) V V \ {(v, v) | v V } of the ratio dT (u, v)/dG (u, v) is minimal.
Here, dG (u, v) refers to the length of a shortest path between u and v in G. The
quantity dT (u, v) denotes the length of the path between u and v in T .
The relation between finding a minimal t-tree spanner of a graph and an MSFCB can be noticed when considering the following unified notation for tree spanner (UNTS) problems. In the UNTS, a problem is defined through a triple
(goal, domain, term) .
Here, goal is either the maximum stretch or the average stretch. Second, as domain,
either all non-tree edges or all edges or all pairs of vertices are considered. Finally,
Introduction
term may be one of the following four: dT (u, v) or dT (u, v)/dG (u, v) or dT (u, v)+w(e)
or dT (u, v)/w(e), with w(e) denoting a weight of an edge. Although not all combinations of goal, domain and term are possible, there remain 20 tree spanner problems,
classified by the UNTS. Interestingly, these 20 notationally different problems collapse to 12 with a general weight function w and to only five, when considering
0/1-weights on the edges.
Among these five problems that do not coincide even in the unweighted case,
arebesides the MSFCB problemprominent optimization problems like the Minimum Average Stretch Spanning Tree Problem [PT01], the Shortest Total Path
Length Spanning Tree Problem [DKP82, WCT00] and the Minimum Diameter
Spanning Tree Problem [HL02].
Generally speaking, the UNTS provides a classification of related problems, which
had not been realized as such before. Hence, interconnections can be revealed and
properties like complexity status or inapproximability factors can be carried forward
between the problems.
The chronology of topics within this introductory part is reflected in the organization of the thesis.
Introduction
which we provide benchmark results for the MSFCB problem on grids which help
evaluating further research.
The MSFCB problem can be interpreted as a problem of finding a spanning
tree that minimizes the sum of path lengths between particular pairs of vertices
in a given graph. Interestingly, finding minimum average stretch tree spanners or
min-max stretch tree spanners of graphs can be interpreted in a very similar way.
In Chapter 3, we provide a classification of several problems that aim at finding
spanning trees in a graph, which minimize the average or the maximum value of
certain distances between particular pairs of vertices in a graph. We propose a
unified notation for these problems, which include several prominent problems in
combinatorial optimization. With this notation at hand, we identify all coincidences
and anti-coincidences of these problems. Moreover, we provide a missing complexity
status for one of the problems and observe that an inapproximability result of one of
the problems can in fact be applied to another problem too, where it had previously
been unknown.
In Chapter 4, the experimental work that is related to the Network Signal Coordination (NSC) problem is presented, thereby coming full circle back to the first
chapter. The experiments conducted are threefold: First, we perform a solver comparison at some example instances for our MIP model. Here, we compare the MIP
solvers CPLEX, MOPS, and SCIP with respect to their ability to find good solutions
in short time. In particular, we run two series of experiments using once the default
MIP solver settings and once settings that emphasize the finding of good primal
solutions. Second, we report on the influence of cycle bases to the computational
behavior of the MIP for the NSC problem. This experiment is of general interest,
because a positive influence of short bases on computation times of mixed-integer
programming formulations of practical applications is expected although very few
studies actually proved it. So, in particular, we investigate the correlation between
the width of a cycle basis and the lower bound obtained by a MIP computation
of 10 seconds. Finally, the third series of experiments is probably the most important one: we evaluate our model by carrying out case studies. Namely, we consider
the real-world inner city networks of Portland and Denver and compare the results
obtained by our optimization approach with results found by other means. For these
comparisons, we use the microsimulation tool VISSIM.
Introduction
A further remark
We assume the reader of this thesis to be familiar with the basic concepts in linear and integer programming, graph theory, and complexity theory. For additional
information on linear and integer programming we refer to [Sch86, NW88]. Good
textbooks on graph theory are for example [Wes96] and [Die00]. Moreover, concepts
in complexity theory that are necessary to follow this thesis are covered by [GJ79]
and [HO02].
As for the parts that deal with traffic engineering concepts or with traffic signal
terms in particular, we refer to Section 1.1.1 for short textual explanations of the
most important terms. Herewith, following most parts of Chapter 1 and Chapter 4
should be unproblematic. Of course, while developing our mixed-linear integer program in Section 1.3, we give formal definitions of all relevant terms, too. However,
additional information can be found in Richtlinien f
ur Lichtsignalanlagen RiLSA,
Lichtzeichenanlagen f
ur den Straenverkehr [ril92] and in the Highway Capacity
Manual [hcm00].
1
THE NETWORK SIGNAL
COORDINATION PROBLEM
In this chapter, we consider the Network Signal Coordination (NSC) problem. The
NSC problem was introduced in 1975 by Gartner et al. [GLG75a], though many
similar optimization tasks were known and have been defined already a lot earlier.
After a brief summary of the most important terms concerning traffic signals in Section 1.1.1, we give a short overview of the most significant optimization problems
on traffic signals in Sec. 1.1.2, also to be able to clearly define the NSC problem
and to restrict it from other problems. Then in Sec. 1.2.1 we formally define the
NSC problem and illuminate similarities to the Periodic Event Scheduling Problem (PESP) in Section 1.2.2. We report on the complexity of the NSC problem in
Section 1.2.3. Thereafter, in Sec. 1.3 we develop in detail a revised mixed-integer
linear programming (MIP) formulation for the NSC problem. Finally, we explain a
possible application of our MIP in practice, see Section 1.4. Parts of this chapter
were published in [MNW06].
1.1
Before we define the Network Signal Coordination Problem, we introduce the most
important terms related to traffic signals and give an overview of what kinds of
optimization tasks concerning traffic signals have been considered so far. Other
surveys on the topic are provided for example by [SS95, tft] or contained in [Lam07].
1.1.1
There is no unique language in the field of traffic engineering in general, and neither
in topics related to traffic signals. Rather, the terms and notation depend on the
7
respective country and language. However, since following this thesis requires only
basic knowledge of traffic terms, in this section we give only a short textual description of the most important terms. Notice that we do not give formal definitions here.
For those, we refer to section 1.3. Nevertheless, whenever it is possible, i.e., when
we do not work with a term, but only want to give an intuition, we omit a formal
definition at all. For complete information see [hcm00] and [ril92].
In traffic engineering one considers traffic, i.e., vehicles, moving through a single
and isolated junction, an arterial, which is a, possibly bi-directionally traversable,
series of junctions, or through a whole network, i.e., an arbitrary set of junctions.
Then, one distinguishes different types of signals at the junctions. Here, the term
signal refers to all signaling devices at a junction. Roughly speaking, the following
two types are the most important ones. On the one hand, there are traffic responsive signals. At these signals, the signal settings react on the present traffic. On
the other hand, fixed-time (controlled) signals do not react on the actual traffic.
Here, after a prescribed time span, called cycle length, the pattern of red phase
and green phase repeats. The particular division of a cycle length into a red and a
green is referred to as (red green) split. At a fixed-time signal, there are usually
different signal groups that control the traffic for particular directions. The green
phases of different signal groups are shifted against each other, since they usually
control competing traffic streams. In addition, the order of the signal groups at
a signal is called phase sequencing. See Fig. 1.3 on page 20 for an example of
a signal timing plan in which the relevant data for one fixed-time traffic signal is
merged.
A very important term is the one of an offset. The (inter node) offset determines how different signals are operated or shifted relatively to each other. That
means the following: at each signal there is a marked out reference point, which
sometimes is the begin of the green phase of the first signal group. Then, the offset
denotes the time span between reference points of two signals at two consecutive
junctions. In this case, there is an offset for each pair of consecutive signals. However, the offset can also be defined for one single signal. Then, it determines the
time span between this signals reference point and a given network-wide zero reference point. See Figure 1.4 for an illustration of both types of offsets. A sketch of
the intra-node offset, which determines the shifting of different signal groups at
one signal, is depicted in Figure 1.5 for example.
Of course, when considering signalized junctions, arterials or networks, several
optimization tasks come to mind. Generally, one is interested in optimizing signal
settings in order to achieve a certain goal. Such signal settings are the red green
split, the phase sequencing, the cycle length, and the offset. As for the goals to
achieve, for example, minimizing the delay or maximizing the bandwidth have to
be mentioned. Here, the term delay refers to the delay that is due to the signalization,
i.e., delay that occurs when vehicles have to wait because of a red. On the other
hand, maximizing bandwidth means that the signals along an arterial or within a
network are adjusted, such that a preferably wide possible corridor through the green
phases of consecutive signals exists, within which the vehicles do not have to stop at
the signals at all. Such a corridor is sometimes called a greenband. See Figure 1.1
Since the introduction of automatic traffic signals in the 1920s, much work and
research has been done on modeling, analyzing, and later also on simulating and
optimizing traffic signals. In this section we mention the most important modeling
and optimization approaches. Notice, however, that we do not claim to provide a
complete overview.
When talking about optimization in the context of traffic signals, one faces many
different optimization tasks. Table 1.1 gives a glimpse of possible differentiations
between them.
One of the first important scientific publications on traffic signals was by Webster [Web58] in 1958. In this pioneering work, he prepared the ground for analyzing single traffic signals, e.g., by providing delay-estimating formulae that are, in a
slightly changed form, still in use today. Using this formulae, Webster also researched
on minimizing the delay by adjusting optimal green proportions at a signal.
Then, during the 1960s research was no more restricted to one single junction, but
rather arterials and networks of signals were considered. In 1963, Newell [New64]
investigated the coordination of signalswith certain assumptions on the density
of trafficalong an arterial of one-directional traffic. In his macroscopic model he
suggests that best coordination is achieved simply by coordinating two consecutive
signals at a time.
Morgan and Little [ML64] and Little [Lit66] then developed an optimization
model that maximizes bandwidth, for the first time using mixed-integer linear programming. In their approach, the authors adjust optimal values to offsets, a common
signal cycle length, and progression speeds, considering a bi-directional arterial of
signals.
In the mid-1960s a first big field-study in coordinating a real traffic network
was carried out in the city of Glasgow, Scotland. In this experiment, conducted
by Hillier [Hil65, Hil66], different types of signal controllings were tested in inner
city sub-network in Glasgow to evaluate their benefit. Later, in 1967, Hillier and
10
Table 1.1: The table provides criteria to distinguish between mathematical approaches for problems dealing with traffic signals. Observe, however, that not all
combinations are reasonable.
Criteria
Possibilities
type of approach
variables
objective
type of signal
type of approach
application on
preconditions on traffic
preconditions on demand
preconditions on signalization
modeling perspective
Rothery [HR67] analyzed the relation between platooning behavior of vehicles and
coordination. In particular, they calculate total delay as a function of the offsets.
For this study the authors investigate four signalized junctions in London, England.
At the same time, a graph theoretical model for optimizing the coordination
in order to minimize the delay was developed by Allsop [All68]. In his article he
proposed an iterative approach that successively expands the considered sub network
for which a solution had already been found.
In 1969 the theoretical work for one of the until now most widely used software
tools for the optimization of coordination was published by Robertson [Rob69]. We
refer to the next section for a more detailed discussion of the properties and attributes
of TRANSYT.
Then, in a series of publications Gartner [Gar72] and Gartner et al. [GLG75a,
GLG75b, GLG76] developed an mixed-integer linear programming approach for network coordination. In fact, the optimization model that we introduce in Section 1.3
is based on their approach.
At the same time, another approach evaluated the possibilities of linear programming for optimizing traffic signal settings. Antoniadis [Ant75], though, did not
include the offset into his model.
A next, also from the theoretic point of view, important approach was the one
by Improta and Sforza [IS82] in 1982. There, the authors developed a mixed-integer
linear program, similar to the one by Gartner et al. [GLG75a], but included a branch
11
and backtrack method that relaxed certain assumptions on the delay functions, which
had been made by Gartner et al.
Dauscha et al. [DMN85], then introduce in 1985 a very general problem framework for finding cyclic schedules of task systems. Nevertheless, their problem formulation is strongly motivated by coordinating traffic signals. In addition, they are
the first who report on the complexity of scheduling periodic events or coordinating
signals, respectively.
In 1989, Serafini and Ukovich provide two important contributions. In [SU89b],
they develop an involved mathematical model specially tailored for the fixed-time
traffic control problem, i.e., for the coordination of fixed-time signals. Second,
in [SU89a] they present a widely noticed general approach that schedules periodic
events, which is, though, very similar to [DMN85].
Then in 1991, a second computer-aided approach, named SCOOT1 , was published
by Robertson and Bretherton [RB91]. However, in contrast to TRANSYT, SCOOT
applies to adaptive traffic signals.
Later, in 1996 Hassin [Has96] presents a flow algorithm approach to the synchronization of networks with an explicit application to the synchronization of fixed-time
traffic signals. The approach consists of a local search heuristic and, moreover, a
characterization of local optima is given. In addition, comparisons to other heuristic
approaches like TRANSYT were carried out on a network in the city of Tel Aviv,
Israel.
A heuristic optimization approach that bases on genetic algorithms was proposed
by Almasri and Friedrich [AF05] in 2005. There, the authors use a cell transmission model which was originally introduced by Daganzo [Dag95]. However, only
adaptively controlled signals are considered.
In 2005, Braun and Weichenmeier [BW05] introduce a second heuristic approach.
In their model, the authors consider several signal settings, including offsets, and
propose a genetic algorithm method. Nevertheless, they report on test runs of the
approach on networks of small sizes only.
In order to show that modeling and optimizing traffic signals indeed attracted
researchers from different fields of expertise we give the following two citations, too:
in 1994 Ianigro [Ian94] developed a traffic model by using petri nets. Then, he
used this model to implement a simulation that finds optimal signal settings for the
considered traffic network. A second interesting approach is the one by Gershenson [Ger05]. In 2005 he considered traffic flow in a signalized network in the context
of self-organizing systems, i.e., he investigates self-organizing traffic lights that adapt
to changing conditions.
Other important contributions that we did not mention in detail before are [Sto68,
CI88]. For an broader overview we refer the reader to [KN03].
1
12
1.1.3
Whereas the last section was intended to give an overview of optimization approaches
to traffic signal related tasks, this section explicitly reflects some software packages
that optimize the coordination of fixed-time signals, both on arterials and within
networks. However, again, we do not claim completeness, but rather mention the to
our knowledge three most important ones: TRANSYT [Rob69], MITROP [GLG76]
and SYNCHRO [syn00].
Remark 1.1. Though nowadays, coordinating traffic signals is done mostly computer aided, still, some techniques are in use, where computers only provide graphical support. For example, especially for arterials, coordinating via a graphic-based
by-hand approach that visualizes greenbands in dependence of offsets is common, see
Figure 1.1.
Figure 1.1: A still today quite usual method to determine a good coordination on
arterials: a graphical approach3 by hand. Notice the greenbands in green and blue
for opposite traffic.
13
into account for the PI. Since in Section 4.3 we work with TRANSYT, we now
give some detailed information on its mode of operation and the adjusted parameter
setting in the following remark.
Remark 1.2. We use version Transyt-7F 10.1 and its genetic algorithm approach
to optimize offsets in a network. The genetic algorithm parameters are: crossover
probability = 40%, mutation probability = 2%, convergence threshold = 0.01%,
number of generations = 100, and population size = 25 which are the defaults.
Further, we calibrated the performance index PI such that it measures delay only.
As the mode we used both the single-cycle mode.
To summarize this paragraph: when developing a model that optimizes the coordination of traffic signals in a network, evaluating it by comparing the results to
the ones of TRANSYT is a must.
In 1976 Gartner, Little and Gabbay introduced the software tool MITROP5
that optimizes traffic signal settings [GLG76, GLG75a, GLG75b]. By then, the
MITROP was one of the first approaches that used integer programming techniques.
In addition, Gartner et al.s model was the first one to simultaneously optimize (one
network wide) cycle length, red-green splits at the signal and offsets. However, since
the relation between these signal settings is quadratic, they use piece-wise linear
approximations. Our approach, which we present in Section 1.3 is actually based on
MITROP. Hence, we adopt most of their notation. Nevertheless, shortcomings of the
MITROP model are the objective function, see Section 1.3.5 for more information,
and the assumption of a global cycle length. However, to the best of our knowledge,
MITROP has not had commercial success, thus, it may be considered a theoretical
approach.
A third software tool, which has to be mentioned when speaking of coordination
of signals in a network, is SYNCHRO [syn00]. Very similar to TRANSYT, the tool
SYNCHRO models all traffic signal settings and searches heuristically for a solution
with small average delay.
However, the three briefly introduced software tools, TRANSYT, MITROP, and
SYNCHRO are by far not the only computer programs that deal with the coordination of signals in networks. Still, we consider them to be the most relevant ones to
compare our approach with.
1.2
In this section we consider the Network Signal Coordination (NSC) problem. First,
in Section 1.2.1 we give a definition of the problem. Thereafter, in Section 1.2.2,
we classify the NSC problem to related problems like the PESP that also deal with
periodically repeated events. Finally, in Section 1.2.3 we prove the NP-completeness
for the NSC problem.
5
14
1.2.1
In this section we will give a mathematical problem formulation for the Network Signal Coordination (NSC) problem. However, the problem formulation will be as what
we want to be understood by network signal coordination. Obviously, there is no
unique possibility to define the problem mathematically. Nevertheless, none of the
already known problem definitions suffices our needs. Either, they are formulated too
general, or, they are formulated too narrow, meaning that many specialized assumptions regarding the signal or the traffic flow, e.g., platoon lengths, are incorporated
into the problem formulation.
So, we try to find a problem formulation that is tailored to the coordination of
traffic signals in networks, while not being over-restrictive with special assumptions,
e.g., on lengths of platoons. Still, we want our problem formulation to map the
following phenomena that are inherent to the practical problem.
In what we refer to as the practical problem we are considering the following
setting. We are given a network of junctions which are signalized with fixed-time
traffic signals. The signals in the network have uniform cycle length. Then, the
objective of the practical problem is to adjust offsets at the signals such that delay
that is due to poor coordination, is minimized. Moreover, we want our problem
formulation to map a macroscopic traffic scenario. Obviously, this is, in general,
a quite restrictive assumption. Nonetheless, under certain conditions, on which we
report in Section 1.3, the macroscopic view is accepted. Then, as a consequence,
we assume the objective function to be separable. This means that for one link, the
delay of the vehicles only depends on the offsets of the two incident signals. Hence,
we may consider each link individually, i.e., the objective is a sum of functions that
evaluate the link performance: so-called link performance functions (LPFs). Finally,
we think it is both, sufficiently exact and sufficiently general, to assume the LPFs to
be continuous and piece-wise linear.
Mathematically, we model the network as a directed graph D = (V, A), where
we refer to an element v V synonymously as junction, signal, node, or vertex. An
element a A we call an arc. We use link or edge synonymously. The network wide
cycle length of the signals is denoted by c with c and the vector V is the
offset vector for the signals.
First, the following observation can be made. In the above requirements, we did
not mention any constraint on an offset v , v V . Hence, any V constitutes
a feasible solution, or can be turned into a feasible solution, respectively, by replacing v with v mod c. Thus, formulating the network signal coordination problem
as a decision problem, which answers the question whether there is a feasible set of
signal offsets does not make any sense.
15
Task:
a=(u,v)A
1.2.2
This section is dedicated to illuminate coincidences between the NSC problem and
related problems such as the Feasible Differential Problem (FDP) and the Periodic
Event Scheduling Problem (PESP).
We are aware that the problem formulation of the NSC problem overlaps with
many known problems. One major difficulty to properly classify the NSC problem
is that different streams of research dealt with very similar problems. For example,
traffic engineers, as summarized in Section 1.1.2, considermostly motivated from
practiceproblems concerning signal coordination, which are formulated very close
to the NSC problem. Nevertheless, there was no uniform mathematical problem
formulation established. Other decisive influences came from the field of operations
research and from computer scientists. In 1985, Dauscha et al. [DMN85] proposed
16
an approach for cyclic schedules for task systems where they put the problem of
coordinating signals in a much more general framework. In addition, it was this
article that reported for the first time on the complexity of a problem closely related
to coordinating signals.
However, it was not until 1989 that Serafini and Ukovich proposed a problem
formulation and a notation that became broadly accepted. In particular, in [SU89b]
they introduced an approach specialized for coordinating signals in networks. Moreover, Serafini and Ukovich published another article in the same year. In [SU89a]
they introduce a mathematical model for periodic scheduling problems. Although
their problem definition is very similar to the one in [DMN85], their formulation of a
Periodic Event Scheduling Problem (PESP) attracted much more attention. The
PESP can be seen as the periodic extension of the so-called Feasible Differential
Problem (FDP).
a w v ua
or decide that no such vector exists.
In both problems, the FDP and the PESP, one considers a directed graph D =
(V, A) and seeks a vector V . This setting is equal to the one for the NSC
problem. However, for the FDP and the PESP, one additionally is given vectors
and u A . With these vectors, constraints on are formulated, non-periodic
ones for the FDP and periodic ones for the PESP. Notice that the absence of such
constraints is the major difference between the NSC problem and these two problems.
However, it is a trivial observation that the c-NSC problem with K = is equivalent
to the PESP with a = 0 and ua = T for all arcs a A.
a (w v )
mod T ua
17
hand, different types of functions are considered to measure the link performance for
the two problems. For the NSC problem, we assume the performance function for
a link to be continuous and piece-wise linear. When optimizing periodic timetables
with the PESP, though, often non-continuous functions that are periodically linear
on an interval of length T , are considered.
Moreover, due to the different origins, different names for the model quantities
have become standard. The period length T of the PESP, for example, is equivalent
to the cycle length c for the NSC problem. Moreover, in the PESP context, the
V , are called potentials, whereas when coordinating signals, they are called node
offsets.
To summarize, the NSC problem and the PESP can be regarded as quite similar.
The differences lie in the bounds on the vector for the PESP and in the different
objective functions. Therefore, modeling or solution approaches to the two problems,
as for example MIP formulations, differ as well, little, but noticeable. However, we
do not go into further detail for that question.
Another interesting aspect when comparing the NSC problem to similar problems
does not become apparent until considering a special model for NSC. It is possible
to model the NSC problem not using variables v for a v V , but exclusively
with variables a := v u for each a = (u, v) A. When doing so, one has
to add certain constraints to the model in order to be able to recalculate the node
offset v out of the link offset a of a solution.6 An important observation is that
such constraints do also appear in the according model for the PESP and, generally,
in every approach that models periodically repeated events using variables on arcs
instead of variables on nodes. Often, these constraints are called cycle constraints,
but, actually, they have had various names like congruence equations [SU89b] or
loop constraints [GLG75a]. In fact, these cycle constraints were already formulated
by Kirchhoff [Kir47] in an aperiodic manner, i.e., for describing properties of voltages
in electrical networks.
Finally, we give a short resume on this section. In the 1950s and 1960s traffic
engineers developed the first models for coordinating traffic signals in networks. Although by then the problems were approached more from the practical side, many
sophisticated mathematical solution methods were proposed. Then, during the years,
the problem of coordinating signals inspires researches from operations research and
computer science. First, they come up with own approaches to signal coordination
and provide first complexity studies. Thereafter, they develop much more general
problem frameworks that cover many different real-world applications like railway
timetabling [Lie06]. Nevertheless, a backfertilization to the actual origin, i.e., the
coordination of traffic signals, could be observed. For example, new mathematical
expertise in network flow theory [Has96] entered the signal coordination approaches.
See Section 1.3.3 for a detailed discussion of these constraints at the example of modeling the
coordination of traffic signals.
18
1.2.3
In 1975, Gartner et al. [GLG75a] introduced an optimization problem that minimizes the delay of vehicles in a network with fixed-time signals by adjusting optimal
offsets. However, although providing a mixed-integer approach for their problem,
which they called Network Coordination Problem, Gartner et al. did not report
on the complexity of the problem.
In 1985, Dauscha et al. [DMN85] proved the NP-completeness of their Cyclic
Schedule Problem, which can be considered to be very close to the problem of coordinating traffic signals. They reduced the Graph k-Colourability Problem, see [GJ79],
to their problem. Later on, other hardness proofs for slightly more general, problems
followed. In 1989, Serafini and Ukovich [SU89a] provided an NP-completeness proof
for the PESP. They reduced the Hamiltonian Circuit Problem, see [GJ79], to their
problem.
In 1996 Hassin [Has96] proves the NP-completeness of his Network synchronization approach by a reduction from the Minimum Cluster Problem, see [GJ79].
The c-NSC problem of Section 1.2.1 obviously belongs to the class NP. Hassin [Has96] reported on the NP-completeness for a variant of his problem which
implies the hardness of the c-NSC problem. Still, in Theorem 1.3 we provide an
alternative reduction from T-PESP proving the c-NSC problem to be NP-complete
for c 3.
Theorem 1.3. The c-NSC problem is NP-complete for c 3.
Proof. It is obvious that the c-NSC problem belongs to NP. We show the NP-hardness
by a reduction from T-PESP, see Section 1.2.2 for a definition, which is known to be
NP-complete for T 3 [Odi94, Odi96].
Let I = (G = (V, A), , u A ) be an instance of T -PESP for some T 3. Define
an instance I = (G = (V , A ), ha , K) for the c-NSC problem in the following way.
First, let G := G, i.e., we take the graph G with its vertex set V and its edge set A.
Further, we set K = 0 and c = T . The functions ha are defined as follows. For
an a A we set
c(ua a )
2ua
2
, a ],
2
c(ua a ) t + c(ua a ) , if t [a
ha (t) =
0,
2
c(ua a )
2ua
c(ua a ) ,
if t [a , ua ],
if t [ua , ua +
(1.1)
c(ua a )
]
2
19
Moreover, the functions ha are defined such that ha (t) = 0 if and only if 0 t
mod c ua a . See Figure 1.2 for an illustration of one function ha .
1
ha (t)
...
...
t
ua
a
0
Figure 1.2: We deduce a function ha using the PESP bounds a and ua , such
that ha (t) = 0 if and only if 0 t mod c ua a .
However, if the instance I is a Yes-instance, then there exists a vector
such that
X
ha (w v ) = 0.
QV
a=(u,v)A
In the previous sections of this chapter, we reviewed existing approaches for the
coordination of traffic signals in networks, and spent some effort in appropriately
defining the Network Signal Coordination (NSC) problem. Now, in this section, a
mixed-integer linear programming approach for the NSC problem is presented.
In this approach, which is based on an approach by Gartner et al. [GLG76],
also [GLG75a] and [GLG75b], we minimize the total delay occurring in a traffic
network by adjusting optimal offsets. In fact, in our MIP model, we also allow
a variable phase sequencing and a non-uniform cycle length at the signals of the
network. Therefore, the MIP approach that we develop in this section actually
solves a more general problem than the c-NSC.
The section is organized as follows: first, in 1.3.1 we introduce the traffic scenario that we consider and discuss the assumptions that we make. Moreover, we
introduce the mathematical notation that we need. Note that we adopt most of the
notation from Gartner et al., see [GLG75a]. Thereafter, in Sections 1.3.2 to 1.3.6 we
20
develop the MIP, which we finally state and discuss in Section 1.3.7. Notice that in
Section 1.3.2 we also motivate the major model technique of using link-wise defined
offsets instead of node-wise defined ones.
1.3.1
Introduction
Figure 1.3: A signal timing plan of Signal v is shown. Exemplarily, the length g
of the green phase for signal group 2, and the length r of the red phase for signal
group 1 are depicted.
A first remark is that we do not consider amber phases at the signals. Instead,
we assume the signal timing plan to only include green and red phases for all signal
21
1.3.2
The Offsets
This section is intended to treat two important tasks. First, we formally introduce
the term offset, in both of its possible interpretations. Second, we motivate, the
decision for the link-wise defined offsets for our mixed-integer linear program.
We begin the development of our mixed-integer linear program with the most
important variables, the offsets. Although in the problem formulation for the NSC
one is interested in offsets for the signals, there are actually two possibilities to define
the offset: node-wise or link-wise. First, we illustrate the difference between the two
viewpoints and, second, we give formal definitions.
In Figure 1.4 the two possibilities to interpret the term offset are shown. First, in
Fig. 1.4(a), we illustrate node-wise defined offsets. There, the term offsets refers to
the time span between the zero point in the signals timing plan and the networks
global zero point. These two reference points can be set arbitrary, however, they
have to be fixed beforehand. Second, in Fig. 1.4(b), we illustrate a link-wise defined
offset. There, the term offset refers to the time span between the beginnings of the
green phases of the upstream and downstream signal of the link. Of course, in both
cases the offset value is only unique modulo the cycle length c.
22
1
2
3
1
2
3
signal zero
point
1
2
3
v
1
2
3
u
(a) node offset
u
(b) link offset
Figure 1.4: The two possible interpretations of the term offset are shown. In (a),
the offset is defined node-wise, whereas in (b) it is defined for each edge a A. In
the example, the link a represents traffic flow between signal group 2 of signal u and
signal group 3 of signal v.
+ pa c = a .
v
u
v
(1.2)
The connection stated in Eq. (1.2) is rather easy to see. We illustrate an example
in Figure 1.5.
Hence, the following can be observed. A model that exclusively contains node
offsets can be considered equivalent to a model that exclusively contains link offsets . Therefore see Equation (1.2). On the one hand, with fixed values of u and v
there always exists a unique pa such that a a with a [ra + a c, ra + a ] exists.
On the other hand, one can uniquely determine the vector out of the link offsets.
However, therefore one node offset has to fixed, which is, though, not a limitation.
This is, because two solutions , V can be considered equivalent if one is just
23
2v
1
2
3
3u
u
1
2
3
a
u
Figure 1.5: The connection between node-wise defined offsets and link-wise defined
ones allows for a straightforward translation between the two terms.
a translation of the other, namely, if there exists a constant q, such that
a = (a + q) mod c,
a A.
(1.3)
However, then, with one v , v V , fixed, one simply uses (1.2) to uniquely determine all v propagating the a , e.g., along a spanning tree of the underlying graph.
That this propagation works out is ensured by so-called cycle constraints, which are
discussed in detail in Section 1.3.3.
Now that we know that there exist several possibilities to model the quantity
offset for our signal coordination approach, we discuss why we decided to exclusively
include the link offsets into our model.
The main reason, why we build up the model with the link offsets, is the objective
function. Using only the node offsets v for the model, we would have faced the
following problem. As we already mentioned, the objective function is separable.
That means, we have an objective function for each arc a A measuring the link
performance. These functions have to be defined in dependence of the difference v
u , for a (u, v) A. However, as it will become clear in Section 1.3.5, where we
discuss the objective function in detail, we need to model the objective, such that it
is convex on an interval of length c, which we know in advance. Using exclusively
the vector, ensuring such a property, in fact, costs a modulo operation per link,
i.e., number of edges many integer variables. As we will see in Section 1.3.3, building
up the model with the link offsets only causes additional constraint and integer
variables, where is the cyclomatic number of the graph and = |A| |V | + 1. Still,
we will remark on the link offset versus node offset question again in Sections 1.5
and 4.4.
So, we decided to use the link-wise defined offset for our MIP approach. However,
we introduce a further variablethe platoons arrival time which is, though,
linear dependent from the offset .
On link a = (u, v) A, the arrival time of the head of the platoon at junction v
is denoted by a . The relation between the arrival time a and the links offset a ,
is given by
a a + ra = a ,
(1.4)
24
with a being the travel time of the platoon and ra denoting the length of the red
phase at the downstream signalgroup of the link. In addition, we require the arrival
time a to lie in the interval [0, c). This will be useful when we use the a for the
objective function, i.e., to evaluate the delay on link a. Note that the bounds for a
imply that for the according link offset, it holds that
a [ra + a c, ra + a ].
(1.5)
In Figure 1.6 we illustrate that connection between the arrival time and the link
offset.
r
c
Figure 1.6: The figure illustrates the simple linear dependency between the arrival
time of a platoon, , and the link offset .
1.3.3
Consider the set L of cycles of G. Note that since we are considering a multigraph,
L includes cycles that contain only two parallel or anti-parallel edges.
For a particular circuit L that is traversed clockwise, F () is the set of forward
edges and R() the set of reverse edges, respectively. With c denoting the networks
cycle length we require the variables to fulfill
X
X
a = n c,
(1.6)
a
aF ()
aR()
Z.
QV
Proof. For simplicity, we assume that all signals have exactly one signal group.
Thus, (1.2), reads as a = v u + pa c. Then, assuming (i), summing up
25
offsets a along a cycle L results in a telescope sum and (ii) holds. On the other
hand, with (ii), one can easily find a V fulfilling (1.2): first, fix one v , for
a v V , to an arbitrary value. Then, second, propagate the link offsets along the
edges of a spanning tree simply using (1.2). Then, the node offsets are well defined
because (ii) holds.
So, the periodicities expressed in Equations (1.6) are physical constraints that
are necessary and sufficient to equivalently model the NSC problem with link offsets,
instead of node offsets. For example when there is a cycle for which (1.6) does not
hold, the transformation of the offsets is not well defined. That means, propagating
the a using Equations (1.2) brings a contradiction. See Fig. 1.7 for a small example.
Note that the intra-node offsets have to be included as well, since the link offsets
A = 65
A = 0
AB = 35
B
CA = 55?
B = 35
BC = 55
C
C = 10
Figure 1.7: Consider this small example network with a cycle length of c = 80s.
Assume that the node offset at vertex A was set to 0. Then propagating the link
offset along the cycle reveals node offsets of 35 and 10 for the vertices B and C,
respectively. However, if then the link offset on link (C, A) is not equal to 70 + 80
but instead, e.g., equal to 55, the node offset at A ceases to be well defined.
do not respect changes of signal groups at a junction. This means the following.
Suppose, we are considering a circuit that contains two links that share a node v.
Further, assume these edges have the same orientation with respect to the clockwise
traversal of the circuit. If then the edges do not share the same signal group at
node v, the shifting between the two signal groups has to be taken into account.
Notice that for the intra-node offsets, too, the traversal direction has to be minded.
For a more evolved example, see Fig. 1.8. There, all relevant time gaps are depicted
through arrows, i.e., directed arcs. Consider now, for example the point in time when
the green phase of the first signal group of Signal 1 begins. Starting from that point,
following the link offset arcs and the intra-node arcs one can traverse a cycle. Thus,
for this example, the sum
21 + 13 23 + 13 23 + 22 + 24 24 + 34 14 31
(1.7)
must equal an integral multiple of the cycle length, since otherwise no consistent
setting of the node offset at Signal 1 is possible. So, Equation (1.6) in fact turns out
to amount to
X
X
X
X
ds(a)
os(a)
ds(a)
a
a +
os(a)
= n c,
u
v
u
v
aF ()
aR()
aF ()
aR()
(1.8)
PSfrag replacemen
26
34
14
13
23
22
31
1
2
3
13
3
4
24
24
1
2
3
21
1
2
3
23
1
1
2
3
Figure 1.8: An example network G = ({1, 2, 3, 4}, {(1, 3), (2, 3), (2, 4), (1, 4)}). In
bluish colors, the link offsets are sketched whereas the intra-node offsets are depicted
in red.
where os(a) and ds(a) again denote the signal groups at the origin and destination
of the link a = (u, v).
An important observation is that, although the cycle equations (1.8) have to hold
for each cycle L, they do not have to be defined explicitly for all cycles. Already
in 1966, Little [Lit66] observed that it suffices to define the cycle equations for the
cycles that are induced by the chords of a spanning tree. Later, in 2002 Liebchen and
Peeters [LP02] introduce the class of integral cycle basis. Moreover, they show that
it suffices to require Equation (1.8) for the elements of an integral cycle basis [LP02]
in order to ensure that it holds for all cycles of the graph. Note that the class of
integral cycle bases is a proper superset of the class of strictly fundamental cycle
bases that are defined by spanning trees [LR07].
A second important observation related to the cycle equations (1.8) concerns the
integer variable n , for L. For computational issues it is important to provide
bounds for them. However, with the help of the bounds on the link offsets, (1.5), one
can easily define an upper bound n and a lower bound n on n . Namely, we get
X
X
X
1
n =
(ra + a )
(ra + a c) +
c
aF ()
(1.9)
aR()
and
1
n =
(ra + a c)
(ra + a ) +
c
aF ()
aR()
X
(1.10)
P
as trivial bounds. Note that the terms are a shortcut form the extended form
as in (1.8). With these bounds on the integer variables at hand, one can consider
the product that was also denoted width of a cycle basis in [Lie06],
Y
n n + 1
27
(1.11)
as quality measure of a basis B for the computation of the MIP. Hence, as we can
choose any integral cycle basis for the MIP formulation, it seems promising to take
the smallest one with respect to (1.11). However, we are only aware of one single
contribution, see Liebchen [Lie06, LPW05], that ever tested the actual influence of
a cycle basis on the computation time of the MIP.
In Sect. 4.1 we provide a second computational study that illuminates the correlation between the quality of a cycle basis and the computational performance of
the according mixed-integer linear program.
1.3.4
In contrast to the approach by Gartner et al. [GLG76], our model offers the possibility
to choose between different red-green split modes, i.e., we allow a variable phase
sequencing. In the previous sections we assumed a fixed signal timing plan at the
signals for clarity reasons. In this section we formally introduce the variable phase
sequencing. This, however, may not be confused with an entirely free choice of
lengths of red and green phases.
In detail, the variable phase sequencing works as follows. For a vertex v V , Rv
denotes the set of signal groups at v. Then, we consider v different predetermined
modes of operation at the signal v,i.e., v different signal timing plans. Nevertheless,
these different signal timing plans must have some properties. First, the signal groups
have to stay the same, i.e., throughout all modes, the turns at a junction that are
controlled by a signalgroup p Rv stay the same. With this, the possibility of an
assignment of the traffic to signalgroups at the upstream and at the downstream
signal is kept.
A second property is that the lengths of the green phases and red phases are
the same for all modes. Thus, the variable phase sequencing offers the functionality of choosing between different modes each giving different intra-node offsets to
v,p
the signalgroups. In particular, the parameter m denotes the intra-node offset at
junction v of signal group p Rv in mode m, where m = 1, . . . , v . Then, we introduce binary variables dv,m for all nodes v and modes m = 1, . . . , v . These variables
manage the selection of a mode. Namely, with Equations (1.12), the actually chosen
intra-node offset of signalgroup p at node v is allocated to the variable v,p :
v
X
m=1
v,p
dv,m m = v,p
v V, p Rv .
(1.12)
Of course, one has to ensure that exactly one mode is adjusted. This is done in
Equations (1.13).
v
X
dv,m = 1
v V.
(1.13)
m=1
28
With Figure 1.9 we give an impression of how such different modes or the according
signal timing plans could look like.
2,3
2,2
1
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
2,1
2,4
4
3
Figure 1.9: An example of different predetermined signal timing plans, i.e., modes,
is shown. Exemplarily, some of the are indicated.
Observe that the introduction of variable timing plans, and thereby changing the
status of the intra-node offsets from parameters, as introduced before, to variables,
is well compatible with the cycle equations proposed in Section 1.3.3. In particular,
ds(a)
os(a)
by the according
and v
in (1.8) one only has to replace the parameter u
u,p
v,p
variables and . Note that here it is important to still be able to assign an
edge to its signal group. However, when considering the definitions (1.9) and (1.10)
of the bounds for the integer variables for the cycles, we observe that these bounds
become much weaker.
1.3.5
In this section we introduce the objective function. Our mathematical model is built
to minimize the total traffic network delay that is due to missing or poor coordination
within the network. Now, we will specify how the delay is actually determined by
evaluating the arrival pattern of the platoons.
First, remember that due to the assumption of the vehicles moving in platoons
through the network, we have a separable objective function. Thus, the delay
on a link does only depend on the offset of that link. Therefore, consider the
link a = (u, v) A and assume certain signalgroups at the origin of the link and at
the destination. Now, the traffic, i.e., the platoon, leaves the junction u. Namely,
throughout the green phase of signalgroup os(a) at the origin signal u, the platoon
is released. Then, the platoon is moving on the link towards signal v. The transit
time on that link is determined by the head, i.e., the first vehicle, of the platoon. See
Figure 1.10(a) for an illustration. Finally, the platoon arrives at signal v. Observe
29
that the length of the platoon then may have changed. This means that although the
platoon was loaded for a timespan with length equal to the length of its green phase
at u, the length of the platoon can have changed while traversing the link. This is
due to dispersion effects on the link that appear especially on links that represent
long streets. However, note that the actual necessary information for calculating the
1
2
3
v
a
2
1
2
3
u
(b)
(a)
Figure 1.10: In Figure (a) we schematically depict the platoon moving from Signalgroup 2 at Signal u to Signalgroup 2 at Signal v. In (b) we focussed on the detail
that is considered when evaluating this links delay. Notice that all information
about intra-node offsets are left out. Only the arrival time, , of the platoon has to
be known.
delay are the arrival time a and the length of the platoon at its arrival at signal v.
The effects that change the platoon shape in between are not pursued.
Further, for the calculation of the delay on link a, we slightly change the perspective, see Figure 1.10(b). Namely, we consider the timespan of length c not as it
is given from the signal timing plan, but beginning with the red phase and followed
by the green phase, as indicated in Figure 1.10(b). This changed viewpoint is also
recognizable in Fig. 1.6.
Then, the situation is as follows. The platoon can pass the signal as long as it
shows green. If the signal shows red, the vehicles have to stop and a waiting queue is
formed. When the signal turns green again, the queue is released. For an illustration
of this queueing see Figure 1.11(a). Now, as we want to quantify the delay, we have
to introduce some parameters. As an input of the model, we are given in advance
the traffic volume on the links. So, with fa indicating the number of vehicles on
link a = (u, v) A, which we measure in vehicles per hour, it is straightforward to
calculate a vehicle arrival rate f pa within the platoon. Namely, it is defined as
f pa =
c fa
.
3600 pa
(1.14)
Notice that we assume a uniform rate with the platoon. So, this rate f pa is the rate
with which the waiting queue is built up, when there is a red at signal v. Then, when
the signal turns green, the queue is released with a saturation rate of sa . However, if
at the beginning of the green phase there are still vehicles arriving, see the situation
that is depicted in Figure 1.11(b), the net-saturation rate is sa f pa .
30
A4
A1
A2 A3
(a)
(b)
(c)
Figure 1.11: Depending on the arrival time of the platoon, the vehicles queue at the
signal. For example, if the platoon arrives at the destination signal at the beginning
of the green phase, (c), and additionally the platoon is not longer than the green
phase, then no delay arises.
Applying these straightforward ideas of how the waiting queue behaves, makes a
calculation of the occurring delay easy. When interpreting this process geometrically,
determining the delay is nothing else but calculating the size of the area that visualizes the queue over time in function of the arrival of the platoon relative to start of
red phase. See for an example Figure 1.11(a) where the arrival time of the platoon
is a = 0. We denote with Z() (z()) the function that measures the total (average)
delay of that link in dependence of the arrival time. Then, we observe the following
total delay for that particular example,7
Z(0) = A1 + A2 + A3 =
p2 f p2a
p2a f pa
+ (ra pa ) pa f pa + a
.
2
2sa
(1.15)
Then, of course, this quantity has to be normalized with the factor pa f pa in order
to obtain the delay per vehicle on link a with a = 0. At this point, we abstain from
providing all possible cases, i.e., explain how Z() is calculated for other values of
and with different assumptions on how, e.g., pa is related to ra and ga . Nevertheless,
an important assumption is the following: we require that the waiting queue must
have dissipated completely when the signal turns red. Without such an assumption,
considering a time span of length c would not be sufficient to determine the delay.
Now, how do we use the function z() to obtain an objective for our mixed-integer
program? A first observation is that we cannot use the above described function z()
as objective function, because it is not linear in the arrival time . In fact, the
function z() is piece-quadratic in which can be seen from Figure 1.11(b). In such
a situation, i.e., the arrival time a lies for example in the interval [ra p2a , ra + p2a ],
the size of the area depicted with A4 is quadratic in a .
As a consequence, we will use a piece-wise linear approximation of z() as follows.
We evaluate in advance the function z() for particular i and use the points (i , z(i ))
for a linear approximation. Note first that z(0) = z(c). Further, it is obvious that
for = 0 the periodic function z() attains its maximum. On the other hand,
for = ra , z() is minimal. Thus, at the best, the platoon arrives exactly when the
signal turns green.
7
assuming pa ra
31
delay
Notice that if pa > ga we do not have a delay of zero even when a = ra . However,
this is not a contradiction to the assumption that the queue must have dissipated
when the signal turns red, e.g. when the net-saturation rate sa ra is high.
ra
ra
(a)
(b)
Figure 1.12: In (a) a link performance function is depicted. The technique to get to
that link performance function is illustrated in (b).
So, for each link we use between 3 and 5 data pairs to define the objective. We
always consider the pairs (0, z(0)), (ra , z(ra )) and (c, z(c)). Additionally, we add up
two more pairs with -value slightly smaller or slightly larger than ra . Then, every
two consecutive data points are used to define a line. See Figure 1.12 for an example.
That means, for a link a = (u, v), let gka () denote the lines, with k = 1, . . . , a ,
where a stands for the number of lines defined for link a. Now, we introduce the
variable za for a link a A for the approximated average delay on link a. Then, we
add the inequalities
za gka (a )
k = 1, . . . , a ,
(1.16)
for all a A to the constraint set of the MIP. Thus, we force the average delay za
to be at least as big as the approximated delay for all sections, which is defined
by the particular line. Of course, if and only if the piece-wise linear function is
convex, Equations (1.16) enforce that za takes a value that really corresponds to
an approximated delay. Notice that this condition on the piece-wise linear delay
function to be convex is the reason why we switched the perspective, as seen in
Fig. 1.10(b). Herewith, we shifted the maxima of the delay function z() to the lower
and to the upper bound on the values, the variable can take.
Finally, the objective function of our MIP is defined by
X
aA
fa za .
(1.17)
32
1.3.6
Non-uniform Cycle-lengths
In this section we discuss how we relax the assumption of a network wide uniform
cycle length at the signals.
Especially when considering medium or large size networks, claiming a uniform
cycle length for the signals of the network may appear restrictive. In a network, not
all junctions are similar with respect to their layout or the amount of different traffic
streams their signal has to cope with. Often, there are signals with a large number of
signal groupsthey control many competing traffic streamswhich therefore need
a large cycle length. On the other hand, for a small junction with, say, just two
competing streams, a short cycle length is possible and, thus, preferable, since less
delay is the direct consequence.
So, a first idea is to introduce variables for the cycle lengths that adjust optimal
cycle lengths at the signals or at least one global variable cycle length for all signals. However, incorporating variable cycle lengths or even just one variable cycle
length into the approach presented so far, is problematic. A major drawback in, for
example, an extension by Gartner et al. [GLG75b] of their approach to the network
coordination [GLG75a], is the definition of the objective, also see Section 1.3.5. With
a uniform cycle length in the network, it was sufficient to approximate the actual
delay with lines. Including a variable global cycle length into the model, however,
would change the delay function Z() to a function Z (, ) of two unknowns: the
arrival time and the cycle length. Thus, one would have to approximate Z (, ) by
planes. Although in [GLG75b], the authors present an approach to do so, we do not
see how the convexity of the approximation of Z (, ) can be ensured for all relevant
cases.
We propose a different possibility to relax the condition on a network wide prescribed uniform cycle length. We allow for arbitrary cycle lengths at the signals,
which, though, have to be given in advance, i.e, they are input parameter and not
variables. Formally, we denote by cv the cycle length of signal v V .
In the remainder of this section, we elaborate on how these non-uniform cycle
lengths are incorporated into the model. A first and general observation is the
following. The quantity lcm({cv | v V }), i.e., the least common multiple of the
cycle lengths, can be seen as a global cycle length for the signalization in the network.
That means, after that amount of time, the signalization pattern of the entire network
repeats. However, this new concept of periodicity can be investigated better when
considering a single edge.
So, consider an edge a = (u, v) A. The signal u operates with a cycle
length of cu , whereas signal v has cycle length cv . Hence, after a timespan of
length lcm(cu , cv ), the signalization pattern for link a repeats. Formally, this is
nothing else but the following observation. For link a = (u, v), the node offsets u
and v induce the same situation, i.e., result in the same delay for the vehicles on
that link, as the offsets u and v , with
u := u + pa gcd(cu , cv )
(1.18)
and pa being an arbitrary integer. Further, with defining the cycle length for a
33
(1.19)
In contrast to the case of uniform cycle lengths, see Equation (1.2), we now face a
problem. Having at hand a solution consisting of link offsets , the node offsets are
not uniquely determined via Equation (1.19). For example in [Hae04] it was shown
that when propagating the link offsets , e.g., along a spanning tree, and thereby
fixing the simply to any possible value, may lead to a contradiction. However, we
report in Section 1.3.7 on how one can calculate the node offsets with the link offsets;
also see [Nac96b].
The next question to discuss is, how the offsets can be coupled to the arrival
time of the platoon. Let a = (u, v) A. Then, lcm(cu , cv ) / cu many different
arrival times at signal vin a period of length lcm(cu , cv )can be observed. The
timespan between these different arrival points is equal to the cycle length cu at
node u. See Figure 1.13 for an illustration. However, it suffices to consider exactly
one arrival time , which then again is denoted by a . In particular, we model the
v
cu
Figure 1.13: The connection between the offsets and the arrival time for nonuniform cycle length case is depicted. In this example we consider an edge a =
(u, v) A. The cycle lengths fulfill 3 cv = 2 cu .
connection between a and the link offset a as
a a + ra = a ,
a A.
(1.20)
Note that this is exactly the same constraint as in the uniform cycle length case, (1.4).
However, we additionally have to give bounds in order to make the relation between
the arrival time a and the offset a uniquely interpretable. Namely, we set a
[0, ca ] and a [a + ra ca , a + ra ]. Notice that this way we allow for exactly one
value for the a variable and one value for the a variable.
Moreover, consider the physical constraints that are modeled by the cycle equations. See Equations (1.8) in Section 1.3.3. For a cycle L, define now the cycle
34
length of , denoted by c as the greatest common divisor of the cycle lengths at the
signals v V that are contained in . Namely, set
c := gcd({cv | v }).
(1.21)
Then, the cycle equations for the uniform cycle length case, (1.8), canonically translate to the non-uniform cycle length case. For an L, we demand
X
aF ()
aR()
a +
X
os(a)
u
aF ()
ds(a)
v
X
aR()
os(a)
u
ds(a)
v
= n c .
(1.22)
Again, similar to Lemma 1.4, the following can be observed for a link offset vector
A : if there exists a vector
A for which Equations (1.19) hold, thenusing
the fact that ca /c is integralalso Equations (1.22) hold for all L. On the
other hand, if (1.22) holds, one can find a vector A for which Equations (1.19)
hold [Hae04]. Moreover, note that the previous fixing of the link offset a to the
interval [a + ra ca , a + ra ] did not constitute a restriction; again because ca /c is
integral.
In the case of uniform cycle lengths it was sufficient to formulate the Cycle
Equations (1.22) for the elements of an arbitrary integral cycle basis, e.g., for the
elements of a strictly fundamental cycle basis induced by an arbitrary spanning tree.
However, when the signals operate with different cycle lengths, not any integral cycle
basis can be used, see [Hae04]. In particular, choosing an arbitrary spanning tree
does not work out in general. On the other hand, Haenelt [Hae04] observes that
there always can be constructed an integral cycle basis B such that if (1.22) holds
for all B, then it holds for any cycle of the graph.
A next observation is that the non-uniform cycle lengths at the signals do not
affect the modeling of variable phase sequences: the Constraints (1.12) and (1.13)
can be adopted.
Finally, let us consider the objective function in the non-uniform cycle lengths
case. With a = (u, v) A, we have to consider a timespan of length lcm(cu , cv ) in
order to determine the delay on link a. See Figure 1.14 for an illustration. Again,
35
cu fa
3600 pa
(1.23)
to determine the traffic ratein vehicles per secondwithin the platoons. Then, the
constraints (1.16) and the objective function (1.17) can be adopted from the uniform
cycle length case. Finally, in Table 1.2 we show the particular platoon lengths that
we used to calculate the functions z() and Z(), see (1.15). These values assume
links with two lanes and are to be adapted when the network layout is different.
Table 1.2: The table shows the grading of the platoon lengths in function of the traffic
volume. One single link is considered and the traffic volume is given in vehicles per
hour.
Platoon lengths p in dependence of traffic volume f
Traffic volume
p in s
0 f < 300
300 f < 600
600 f < 900
14
15
17
Traffic volume
0900 f < 1200
1200 f < 1500
1500 f
p in s
19
21
23
We conclude this section giving some remarks. A first observation is that adding
the functionality of non-uniform cycle lengths to the model, does not increase the
model in terms of number of constraints and number of variables. The preprocessing
step to define the linear functions gka () for the objective becomes more involved,
which is, though, computational negligible.
Furthermore, it has to be mentioned that in the non-uniform cycle lengths case,
it is more difficult to provide piece-wise linear functions that are convex. Remember
that the convexity for the link performance functions is essential for the model. The
reason for this difficulty is the following. In the uniform cycle lengths case it was
obvious thatfor a link a Athe link performance function takes its maxima at
the according a s lower and upper bound, 0 and c, respectively. Further, a minimum
was attained at a = ra . In case of non-uniform cycle lengths in the network, maxima
and minima of the link performance functions cannot be located immediately.
Although, theoretically, there is no reason to restrict the values for the cycle
lengths, one should keep the quantity
lcm({cv | v V })
gcd({cv | v V })
(1.24)
Assume for this example that we consider integer cycle lengths and offsets.
36
In this section we finally state the mixed-integer linear program for the NSC problem
in which we optimize the coordination of fixed-time traffic signals. The MIP contains
the variables and constraints that we developed in the Sections 1.3.1 to 1.3.6. In the
remainder of this thesis we refer to this MIP as NSC MIP or simply by (1.25):
X
(1.25a)
min
fa za
aA
aF ()
a +
aR()
v
X
= n c
B,
(1.25b)
v V, p Rv ,
(1.25c)
v V,
(1.25d)
a A, k = 1, . . . , a ,
(1.25e)
B,
(1.25g)
v,p
dv,m m = v,p
m=1
v
X
dv,m = 1
m=1
za gka (a )
a a + ra = a
n
Here,
Z, a
n n n
a=(u,v)F ()
a A,
a=(u,v)R()
(1.25f)
(1.25h)
u,os(a) v,ds(a) .
To summarize: Equations (1.25b) are the cycle equations on the offsets. Remember that B is an integral cycle basis: an arbitrary one in the case of uniform
cycle lengths at the signals and a specifically constructed one, see [Hae04], otherwise.
37
Further, Equations (1.25c) and (1.25d) manage the selection of a mode for each signal, i.e., the selection of one particular prescribed signal timing plan. Moreover,
Equations (1.25f) couple the offsets with the arrival times , which are thenvia
Inequalities (1.25e)used to determine the average delay z for a link. Finally, (1.25g)
and (1.25h) formulate bounds, cf. (1.9) and (1.10), and the integrality of the cycle
variables.
However, after having solved (1.25), an important task remains. A solution
of (1.25) contains, among others, a set of offsets a and arrival times a for each a
A. Nevertheless, as we want to evaluate the MIP, we are interested in node offsets.
Hence, one has to carry out a backtransformation of the link offsets a to node
offsets v . For describing how such a backtransformation works we distinguish two
cases.
First, assume that there is a uniform cycle length at the signals, i.e., cu = cv
for any u, v V . Then, backtransformation is done simply by propagating the link
offsets along the edges of a spanning tree T of G where one node offset was fixed
before. This is done using Equation (1.2). The key observation to see that this works
is that the cycle constraints (1.25b) ensure that (1.2) holds for the non-tree edges
of T . Of course, for that propagation, intra-node offsets have to be considered.
On the other hand, when facing different cycle lengths at the networks signals,
backtransformation is more complicated. In general, propagating the link offsets
along a spanning tree is not possible anymore, because with (1.19), the node offsets
are not uniquely determined by the link offsets a . Instead, one has to solve the
mixed diophantine linear equation system consisting of (1.19) for all a = (u, v) A.
This can be done algebraically or by solving a separated MIP. To the best of our
knowledge, there is no efficient combinatorial approach for the backtransformation
of the link offsets to node offsets.
In Section 1.5 we discuss possible enhancements of the model. Also notice that
in Chapter 4 we conduct experiments related to the NSC MIP (1.25). In particular, we will evaluate the MIP at some real-world example networks. Moreover, we
test the correlation between cycle bases used for the MIP formulation and the MIP
performance. In addition, we use the real-world instances to conduct a MIP solver
comparison.
1.4
In this section we address the issues of an NSC model in general and the NSC
MIP (1.25) in particular.
The popularity of signal timing optimizers, such as TRANSYT [Rob69] and SYNCHRO [syn00], lends practical relevance to the network coordination problem. Such
optimizers accept as input a description of the supply, i.e. the street network in the
study area (mainly links with travel times, permitted maneuvers at junctions), and
of the demand, i.e. traffic flows through the network. The demand can be either
observed or derived from a demand model. Interestingly, demand models are themselves extending in the direction of traffic engineering, supporting models for signal
38
control at junctions and offering analysis methods for them, e.g. by incorporating
capacity analysis according to the HCM [hcm00]. These models aim to provide a
more realistic node impedance for route choice by incorporating capacity analysis
into the assignment step. So far, most packages are limited to the optimization of
cycle lengths c and green time fractions (g/c), as run time requirements have precluded network coordination. This is due to the nature of the solution methods often
used in practice (genetic algorithms or quasi-exhaustive search). A faster solution
method would enable an even closer integration between macroscopic demand modeling and network coordination. The Network coordination could then be part of the
assignment process, allowing to update route flows in response to changed offsets,
and thus reach an equilibrium between demand and supply, including all aspects of
signal control.
Figure 1.15: The organization of the overall optimization process with components
assignment, local optimization and network coordination. We focus on the latter one
stressing the requirements in the context of the optimization scheme.
This overall optimization process is illustrated by Fig. 1.15. Given OD flows
(from previous model steps) are assigned to a network with signal control, using
an assignment method which calculates node impedance from signal timings and
offsets between different junctions. The assignment produces node flows (traffic
volumes per maneuver at each junction) which are input to the local optimization
of g/c. Path-based assignment procedures also produce route flows on the subpaths between successive signalized junctions, or more specifically, between successive
signal groups or stages. In addition the assignment yields the volumes of non-platoon
traffic arriving at a signalized junction from non-priority un-signalized maneuvers
upstream. A network coordination optimizer takes these sub-path flows and globally
optimizes the offsets. Then, the set of offsets are fed back to assignment. The loop
is executed at least once and continues until route impedances (including loss times
experienced at signal controls) and flows converge.
However, before the application of the network coordination presented in this
39
section can actually be realized in practice, some open issues have to be discussed.
A first question is how the coordination of the signals is actually distinguishable
within the assignment step, see Figure 1.15. Therefore, one has to develop traffic
assignment models that include offsets of signals into their impedance functions.
A second important issue is the one of the convergence of this iterative procedure
illustrated in Fig. 1.15. It must be assumed that any strict claim on convergence
behavior will not be easy to prove.
However, let us remark that an application of network signal coordination in
general and the MIP (1.25) in particular within the iterative procedure, Figure 1.15,
is not the only possible application. Rather, the optimization of the coordination of
signals in a network is an important task in itself, i.e., the usage as a stand alone
tool is a relevant application, too.
Depending on the respective application, different requirements on the models
computational tractability arise. Whereas for the application of a network signal
coordination approach computation times in the timescale of seconds are required,
practitioners consider computation times of several hours for a stand alone application to be reasonable.
1.5
In this section we considered the Network Signal Coordination (NSC) problem. After
a short introduction of the most important terms related to traffic signals, we briefly
reviewed previous work on coordinating traffic signals. Then, in Section 1.2.1 we gave
a formal mathematical problem definition of the NSC problem. For this definition we
ensured that certain phenomena of the problem in practice are covered. Thereafter,
we investigated connections between the NSC problem and related problems such as
the Periodic Event Scheduling Problem and stated an NP-completeness proof for
the NSC problem. Then, during Section 1.3, we derived in detail a revised MIP
formulation for the NSC problem. Finally, we reported on possible applications for
a network signal coordination model.
In the remainder, we discuss open questions and tasks related to the NSC problem
in general, and our MIP approach in particular.
We decided to build a model for the c-NSC that uses offset variables on the edges
instead of node offset variables. The reason for doing so was the objective function.
Nevertheless, it would be interesting, whether a MIP approach that together with
the link offsets involves node offsets, performs comparable. Experiences from the
PESP context, show that MIP approaches with variables for the nodes sometimes
computationally behave well.
When considering our NSC MIP (1.25) several possible extensions come to mind.
For example, variable cycle lengths, variable lengths of red phases and green phases
and even variable travel speeds for the vehicles are topics for further research. However, incorporating these additional functionalities, would cause major changes of
the model and is not at all trivial to accomplish.
40
Moreover, an important question when building up a model for network coordination, is, first, the treatment of signals that are not predestined for the optimization
of offsets, e.g., traffic responsive signals, and second, the treatment of unsignalized
junctions. This remains an issue, since in almost all inner-city traffic networks of relevant size, one can find unsignalized junctions and traffic responsive signals together
with fixed-time signals. In this very same context, the treatment of public transport
means, which often actuate signal controllers, is to be mentioned as an open task.
However, when considering the possible application of network coordination of
Section 1.4, one observes a limitation of the network coordination approach. A
question that arises is why should one separate the network coordination from the
traffic assignment?. A merging of these two optimization tasks can be motivated
from both sides. First, from the viewpoint of network coordination optimization,
it is not entirely realistic to assume that the vehicles move on prescribed paths
through the network. Rather, it can be assumed that for example a coordination
that penalizes some paths, actually changes the drivers route choice. Second, from
the viewpoint of traffic assignment, it is unrealistic not to consider signal settings
when defining node impedance functions. And, of course, the coordination of signals
indeed is an important signal setting.
Although the idea of an integrated model of traffic assignment and signal coordination is, of course, not new, so far, only approaches merging the traffic assignment
with models concerning traffic signal settings different from coordination, were presented. Namely, a main difficulty seems to lie in incorporating non-local quantities
like the coordination into an traffic assignment approach.
Thus, since a fully integrated model seems rather complicated at first sight, one
could think of the following. A first step could be including the offsets as parameters into an traffic assignment model, which is necessary to process the iterative
optimization scheme (Figure 1.15). That means investigating and quantifying the
influence of a coordination of traffic signals to travel times and route choice of drivers
in networks.
2
STRICTLY FUNDAMENTAL CYCLE
BASES ON GRIDS
In this chapter we are going to investigate strictly fundamental cycle bases on square
planar grid graphs. We develop both new lower and upper bounds for the problem.
Moreover, we conduct some computational experiments that improve on the experimental studies carried out so far. This chapter is based on [KLRW08] and [LWK+ 07].
2.1
Introduction
42
several ways. First, planar grids are almost regular (more than n 4 n vertices
have degree four) and within a fixed distance, the subgraphs around almost each
vertex are isomorphic. Hence, any heuristic that bases important decisions on local
2.1 Introduction
43
configurations is likely to perform poorly. Second, if G was a tree, then in the MSFCB
problem no decisions are to be made and the problem clearly becomes trivial. An
appropriate measure for the tree-alikeness of a graph is its treewidth [Bod93]. With
respect to that measure, grid graphshaving (n) edges and treewidth nare
prominent examples of being far away from being a tree [RS86]. Thus the MSFCB
problem is likely to maintain its combinatorial hardness when considered on square
planar grids. Finally, also from a practical perspective planar grids are very suitable.
Many of the relevant instances in several applications are planar graphs or even
planar grids (e.g., electrical engineering and traffic light scheduling).
Focusing on grid graphs could appear narrow. But it is commonly believed that
these hold the key to better algorithms for cycle bases. Indeed, for square planar grid
graphs Alon, Karp, Peleg, and West [AKPW95] designed spanning trees that induce
cycle bases of length 43 n log2 n + O(n), see Section 2.4. They proved that these trees
are asymptotically optimal.
Using this asymptotic upper bound we demonstrate how heuristics that base
major decisions on local configurations risk failure. For instance, the degree-based
C-order heuristic [LAM05] can be implemented to compute so-called Machete-trees
(cf. Section 2.4 and Figure 2.10 for example trees). Although these trees minimize
3
the maximum stretch, they yield a poor MSFCB objective value of only (n 2 ). In
particular, such heuristics may miss the optimum by a non-constant factor. This
again underlines the property of grids being a relevant testbed.
In fact, Liberti et al. [LAM05] also select grid graphs as one of their testbeds. On
the 50 50 grid they observe that their new C-order heuristic attains an objective
value of 46,452, showing that the value 48,254 of the NT heuristic [DKP95] is even
worse. Unfortunately, such an isolated comparison does not clarify whether these
are indeed good absolute objective values. Amaldi et al. [ALMM04] consider grid
graphs in their computations as well. They report a solution of objective value 23,026
for the same grid-size, and it was obtained by local search techniques. Liebchen
et al. [LWK+ 07] further improved this value to 21,920 and is it not clear which further
improvements are possible. In other words, there is a need for good benchmark values
for the MSFCB problem for the particularly challenging case of planar grid graphs
also for the future evaluation of new heuristics.
Contribution. Alon et al. [AKPW95] established that an optimal solution for the
MSFCB problem on grid graphs is of length (n log n), with n beeing the number
ln 2
of vertices in the graph. In detail, they proved 2,048
n log2 n O(n) to be a valid
4
1
lower bound and proposed trees that achieve 3 n log2 n + O(n). In this chapter we
improve on both of these bounds. Namely, in Section 2.3.1 we introduce a lower
1
bound of 12
n log2 n O(n) whereas in Section 2.4.1 we provide a family of spanning
trees with length no more than 0.979 n log2 n + O(n).
However, when we consider small grids this new asymptotical lower bound is not
applicable. A lower bound for the MSFCB problem that is not restricted to planar
grids is the length of a general minimum cycle basis, and it was used for example by
1
44
Amaldi et al. [ALMM04]. Applying this MCB bound to square planar grids yields a
of grid graphs to come up with a new lower bound of 6.25n 23.5 n + 34, which
constitutes the best known lower bound for small grids. However, this bound is not
tight for 8 8 grids. Hence, in Section 2.3.5 we present a combinatorial proof for a
tight lower bound for the 8 8 grid.
Furthermore, in Section 2.3.3 we develop additional valid inequalities to a mixedinteger linear programming approach for the MSFCB for general graphs by Amaldi
et al. [ALMM04]. In addition, we present a new MIP model in Section 2.3.4 that is
tailored for square planar grids.
Then, in Section 2.5 we conduct experiments and provide benchmarks for upper
and lower bounds for the MSFCB problem on grids up to the size of 100100. These
results improve on previous ones by [ALMM04] and [LAM05].
Finally, we present a gallery of best known spanning trees for sizes 9 9 up
to 20 20.
2.2
Prelimenaries
We consider cycle bases of a 2-connected simple undirected graph G = (V, E). Define
n = |V |, m = |E|, and = m n + 1, where is the cyclomatic number of G. Let
C be a circuit (cf. [Sch03, Ch. 3]) in G and denote by C its {0, 1}-incidence vector.
The cycle space C of G is the following vector subspace over GF(2):
C := span ({C | C circuit in G}) .
A cycle basis B of G is a set of circuits of G whose incidence vectors are a basis
of C. The length (B) of a cycle basis of an unweighted graph is defined as (B) :=
P
CB |C|. A minimum cycle basis (MCB) of a graph G is a cycle basis of G of
minimum length.
A set of circuits {C1 , . . . , C } such that
Ci \ (C1 Ci1 ) 6= ,
i = 2, . . . ,
is clearly a cycle basis. We call such a basis weakly fundamental. Notice that these
were already considered by Whitney [Whi35] in 1935.
Let T be some spanning tree of G. Depending on the context, we either regard T
as a subgraph of G or as a set of edges T E. For e E \ T , we denote by CT (e)
or Ce for shortthe fundamental circuit that e induces with respect to T , i.e., the
unique circuit in T {e}. There are fundamental circuits associated with T .
These form a cycle basis which is called strictly fundamental. Here, we may write
(T ) instead of (B). A minimum strictly fundamental cycle basis (MSFCB) has
minimum length among the set of strictly fundamental cycle bases.
In general, strictly fundamental cycle bases are a proper subset of weakly fundamental cycle bases, which in turn are a proper subset of general cycle bases of
undirected graphs. In general none of the three corresponding minimization problems coincide [LR07].
45
At the same time, given a spanning tree T of G and any edge f T , the graph
Tf := T \ {f } is a forest comprising precisely two trees with vertex sets Sf and S f
respectively. We denote by (Sf ) the set of edges in E with precisely one end-vertex
in Sf . This set (Sf ) is called the fundamental cut of f with respect to T . There
are n 1 fundamental cuts associated with T . These form a cut basis
P (or co-cycle
basis) which is called strictly fundamental. We denote by (T ) := f T |(Sf )| the
length of this strictly fundamental cut basis.
With N N, the planar grid graph GN,N is the graph on V = {1, . . . , N }
{1, . . . , N } with
E = {{(i, j), (i , j )} | |i i | + |j j | = 1} = {{u, v} | ||u v||1 = 1}.
In a graphical representation, e.g., in an embedding into Z2 , the first index of a
vertex represents its x-coordinate, the second index its y-coordinate. Further define
the vertical distance of an edge e = {(i, j), (i , j )} from a vertex u = (a, b) as
distV (e, u) := max{j, j , b} min{j, j , b}.
The horizontal distance is defined analogously using the first coordinates i, i , a.
The grid graph GN,N has n = N 2 vertices and contains m = 2 N (N 1) edges.
Its cyclomatic number is (N 1)2 . During the whole chapter we refer to N as
the dimension of the grid. We collect some well-known simple properties of the cycle
space of such grids in the following proposition.
Proposition 2.1. The planar grid graph GN,N has a unique minimum cycle basis B.
In B each basic circuit contains precisely four edges, thus (B) = 4 = (n). The
basis B is weakly fundamental. But for N 4, B is not strictly fundamental.
2.3
Lower Bounds
In this section we investigate lower bounds for the MSFCB problem on unweighted
planar grid graphs. As mentioned in the introduction Alon et al. [AKPW95] proved
that an optimal solution is of size (n log n). However, a detailed view on their
ln 2
presentation reveals that they achieve a lower bound of 2,048
n log2 n O(n). In
Section 2.3.1 we improve this lower bound by a factor of approx. 245 reaching a
1
lower bound of 12
n log2 n O(n). Not surprisingly, this lower bounds only works
out for very large dimensions. Thus, in Section 2.3.2 we developed a combinatorial
lower bound of 6.25n 23.5 n + 34 which outperforms the asymptotical bound for
dimensions up to N = 218 . Further, we present two different mixed-integer linear
programming approaches. In particular, in Section 2.3.3 we introduce an approach by
Amaldi et al. [ALMM04] for general graphs whereas in Section 2.3.4 we develop a MIP
that is purpose-built for grid graphs. We conclude the section on lower bounds with
presenting a combinatorial approach for a tight lower bound for dimension N = 8
derived from ideas from Section 2.3.2.
46
2.3.1
47
H
an edge f in Pu,v as
pos(f
H
, Pu,v
)
1,
pos(f , PwH2 ,v ) + 1,
if f = f,
otherwise, i.e., f PwH2 ,v .
V . Observe that in genAn equivalent procedure defines the vertical pseudo-path Pu,v
H P V 6= P ; for an example of such a path P , in Figure 2.1(a) consider the
eral Pu,v
u,v
two vertices with Cartesian coordinates u = (1, 1) and v = (2, 2).
As an example for a pseudo-path, consider the dual graph T in Figure 2.1(a)
and the black vertex u in the center of the grid. Let v be the penultimate vertex
V exactly consists of the black edges, highlighted
of the u, F -path. With this, Pu,v
in Figure 2.1(e). In addition, the position of an edge is tightly related to the quanV ) = dist (f, u), for an edge f that is contained in a
tity dist. Namely, pos(f, Pu,v
V
vertical pseudo-path starting at u.
Lemma 2.2. Let u = (i, j) V (G ) \ {F } be some vertex in the dual grid and let
v be a vertex on the (unique) path P between u and F in T . Further, let Pu,v P
be a pseudo-path between u and v. Then, the sizes of the fundamental cuts can be
bounded by
|(Sf )| 2 pos(f, Pu,v ), f Pu,v .
(2.1)
Proof. Without loss of generality, regard Pu,v as a horizontal pseudo-path, and assume v = (i , j ) where i = i + |Pu,v |.
48
Consider some edge f Pu,v and the induced set Sf V (G ) such that u Sf
and F S f , where (Sf ) is the corresponding fundamental cut. As we consider
a horizontal pseudo-path the y-coordinate of both vertices of f is equal and their
x-coordinates are i + pos(f, Pu,v ) 1 and i + pos(f, Pu,v ), respectively. Remember
that Pu,v is contained in the unique u, F -path P of T . Therefore, all vertices
between u and f are contained in Sf . In particular, for each integer with i <
i + pos(f, Pu,v ), there exists a vertex in Sf with as x-coordinate.
Out of those vertices in Sf with x-coordinate consider the vertex wmax (wmin )
with maximal (minimal) y-coordinate. Note that wmax and wmin may coincide. Now,
to wmax (wmin ) one edge in the cut (Sf ) can be assigned, because the dual vertex
directly above (below)possibly F is not included in Sf . Hence, for each we
get a contribution of two distinct edges and therefore a lower bound of 2 pos(f, Pu,v )
on |(Sf )| in total.
Remark 2.3. In the proof of Lemma 2.2 we detect special edges to be contained
in the cut. However, all of those edges are vertical edges; recall that we assume
w.l.o.g. that Pu,v is a horizontal pseudo-path. Thus, one can add a +2 to the right
hand side in (2.1) since every fundamental cut contains at least 2 edges of both
adjustments, vertical and horizontal. On the other hand, this is of no effect for the
n log n-coefficient.
Note that in general (2.1) does not hold when choosing a vertex v which is not
contained in the unique u, F -path in the dual tree. Furthermore, (2.1) does not
hold either when considering all the edges of an ordinary path P instead of one of
its two pseudo-paths. Moreover, the estimate in (2.1) can be far from being tight.
Consider in Figure 2.1(f) the vertex u having Cartesian coordinates (16, 2). In a
vertical pseudo-path that starts at u the first edge only contributes 2, although it
induces a cut of length 18.
In order to employ this powerful tool for estimating sizes of cuts, we need some
more definitions. An important concept for our approach is the distance between
two dual vertices. Let the grid graph G \ {F } be embedded in Z2 as described in
Section 2.2 and let u = (i, j) and v = (i , j ) be two vertices of it. Then the distance
du,v is defined as max{|i i |, |j j |}, or ||u v|| . It is a simple observation that
for any two distinct vertices u, v that are connected by a path in T \ {F } at least
one of the two pseudo-paths from u to v has precisely du,v edges.
Next we assign tags to vertices in order to organize
them
in so-called levels. In
2
k
k
a dual grid G \ {F } with (N 1) = 2 + 1 2 + 1 vertices we establish k
different levels of vertices as follows. The level k only contains the unique grids
center vertex. The center-vertices of the four quarters of G \ {F } (which overlap
on their borders) constitute level k 1. Recursively, each of these four quarters is
again subdivided into four new quarters whose centers define the next levels. Hence,
for 1 k there exist 4k level- vertices in G \ {F }.
We further assign boxes to level-vertices. These boxes are exactly the quarters
which were used to define their center-vertices as belonging to a certain level
technically, for a vertex u of level with u = (i, j) V (G ) \ {F }, we define
49
1 . Further, we call the set
its
set
of dual vertices Bu = v | du,v 2
box as the1
the border of Bu .
v | du,v = 2
We illustrate the arrangement of the levels in Figure 2.1(a). There, the 64 = 441
level-1 vertices are marked as small light-grey circles. With increasing level index the
level-vertices are sketched with increasing intensity culminating with the one level-4
vertex in the grids center. In Figure 2.1(b)-2.1(d), the boxes of the level-vertices are
indicated by thin lines. In these figures levels 1, 2 and 3 are marked. In Figure 2.1(e),
the box of the level-4 vertex constitutes the whole dual graph, except for F .
We count along the following pseudo-paths. Every level-vertex u serves as a
starting point of one pseudo-path. We consider the unique u, F -path P in G .
Every such path has to intersect the border of box Bu . Let v be the first such border
vertex. For every level-vertex u we denote by Pu,v the longer one of the two pseudopaths from u to v. Consequently, if u is a level- vertex, then |Pu,v | = 21 = du,v .
Sometimes, for a level-vertex u we may write Pu instead of Pu,v , as a shorthand.
Lemma 2.2 suggests that we can count for every edge e Pu a contribution of
2 pos(e, Pu ) to the global lower bound. However, as pseudo-paths of different levels
may overlap (cf. Figure 2.1(c)) this may over-estimate the lower bound. In fact,
there even exist spanning trees such that one edge is contained in a pseudo-path of
every single level. In the next two subsections we propose two strategies to overcome
the inconvenience of overlapping pseudo-paths.
The following two lemmas are the key observations to justify this approach.
Proposition 2.4. Let u 6= u be two level-vertices of levels and , respectively,
such that their pseudo-paths Pu and Pu share some edge e. Then 6= .
Proof. The claim follows from two facts. First, every pseudo-path only consists
of edges within its box. Second, boxes of the same level only intersect at their
borders.
Lemma 2.5. Let , be two levels. Let u 6= u be two corresponding level-vertices
such that their pseudo-paths Pu and Pu share some edge e. By Proposition 2.4, we
may assume that < . Then
pos(e, Pu ) 2 pos(e, Pu ).
(2.2)
50
(b) Level 1
(c) Level 2
(d) Level 3
(e) Level 4
(f)
indicate the connections to F . In (b)-(e) one can see which parts of T are used for
bounding the corresponding cut-lengths in each level-iteration. Those are depicted
using black lines, whereas the grey parts stand for pseudo-paths of previous levels.
Note that the pseudo-paths do not necessarily start from the vertex they belong
to. In addition, the boxes of the vertices of each iteration are illustrated. In (f) we
illustrate what a small part of the tree is actually taken into consideration to obtain
the desired lower bound. The bold edges are the ones that are shared by the level-4
pseudo-path and some pseudo-path of level 1, 2, 3.
51
s1
X
pos(e, P i ).
(2.3)
i=1
Now we continue with presenting two ways to deal with overlapping pseudo-paths.
1
In a first approach, which then gives the 16
n log2 nO(n) lower bound, we voluntarily
count less for each occurrence of an edge on some pseudo-path. This way, we may
eventually sum over every occurrence of an edge on some pseudo-path. Thereafter,
in an alternative refined analysis, we maintain some bookkeeping in which we keep
track of all possible occurrences of the edges on pseudo-paths.
So, recall from Lemma 2.2 that for every edge e twice its position on the path
of the maximal level s that e occurs on is a valid lower bound for (Se ). By Corollary 2.6 we have a valid lower bound when summing over every pseudo-path P i that
the edge e occurs on, its position values pos(e, P i ), i.e.,
(2.1)
(2.3)
|(Se )| 2 pos(e, P )
s
X
pos(e, P i ).
(2.4)
i=1
Theorem 2.7. Let GN,N be the planar grid graph with n = N 2 = (2k + 2)2 vertices.
For every spanning tree T of GN,N the size of the strictly fundamental cycle basis
induced by T can be bounded from below by
1
n log2 n + (n).
16
Proof. Let E(P) be the set of edges that appear on some pseudo-path corresponding
to the dual tree T of T . Then, using (T ) = (T ), we conclude
(T ) =
eE(T )
|(Se )|
(2.4)
eE(P)
k
X
=1
|(Se )| =
X
Pu
u level- vertex
k
X
=1
ePu
Pu
ePu
u level- vertex max. level for e
pos(e, Pu ).
|(Se )|
52
1
k
2X
k
X
X
1
2
k
k
4 +2
(T )
4
i =
4
8
i=1
=1
=1
1
1 k
1 k
=
4 k+
4 2k = (N 2)2 log2 (N 2) + (N 2 )
8
4
8
1
n log2 n + (n).
=
16
In the remainder of this section we perform a refined analysis using our concept
of counting along pseudo-paths showing that, in fact,
(T )
1
n log n + (n).
12
In order to obtain a simple asymptotic proof of the n log n lower bound as presented above we tried not to over-estimate contributions of edges. Now, we will show
that we do not have to abandon the factor of 2 (cf. Lemma 2.2) as we did before
in (2.4). Of course, this requires a more detailed examination of the occurrences of
edges in different pseudo-paths.
The approach in the sequel is as follows. For a given tree T we investigate occurrences of edges in pseudo-paths. Notice that we are interested in the highest level
that an edge occurs on; see (2.4). However, for the analysis we also need low-level
occurrences of an edge. We perform this level-examination using Algorithm 1. Nevertheless, the algorithm only does a bookkeeping job; the tricky part of the section
will be the analysis of the output b of the algorithm. Roughly speaking, Algorithm 1 works as follows. We iterate over the different levels, over the corresponding
level-vertices, and over the edges on pseudo-paths that lie within such a level-vertex
box. Then, in the core of the algorithm, Lines 510, it is checked whether an edge
is contained in the pseudo-path Pu of the box Bu . If the answer is positive we
store in (, ) a lower bound on the size of the cut induced by the current edge,
cf. Lemma 2.2. If the answer is negative, the former lower bound is carried over,
Line 8. For the upcoming analysis, we also keep a quantity (, ) that measures the
marginal increase of (, ) when having incremented the level.
Lemma 2.8. Let 1 0 k. Then after the 0 -th iteration of Algorithm 1
|(Se )|
0
X
j=1
(e, j),
(2.5)
Proof. First observe that by the definition of (e, ) in Algorithm 1 the sum in (2.5)
is in fact a telescoping sum, and thus can be simplified to
0
X
j=1
(e, j) = (e, 0 ).
53
init (e, 0) = 0, b = 0
for = 1, . . . , k do
for all level- vertices u do
for all e Bu E(P) do
if e Pu then
(e, ) 2 pos(e, Pu )
else
(e, ) (e, 1)
7
8
9
10
eE\T
|(Se )|
=
eE(P)
k
X
k
X
u level
vertex
k
X
=1
=1
=1
Lem.2.8
|(Se )|
k
X
(e, )
eE(P) =1
(e, )
(2.6)
u level
vertex
ePu
(2.7)
u level
vertex
ePu
"
ePu
(e, )
ePu
(e, 1)
!#
(2.8)
Here, Equation (2.6) is indeed the result of re-ordering edges of pseudo-paths, taking
into consideration that all summands with (, ) = 0 are disregarded. Recall that
by the definition of the boxes, two pseudo-paths of the same level do not intersect.
In (2.7) we used the definition of (e, ); see Line 9 of Algorithm 1. Now, with (2.8)
where we just regroup the previous summation we continue as follows. Separately,
54
in Lemma 2.10 we give a lower bound on the positive term in (2.8) and identify in
Lemma 2.11 an upper bound on the negative term in (2.8).
Lemma 2.10. Let u be a level- vertex and let Pu be its associated pseudo-path.
Then
X
1
(e, ) = 4 + 21 .
4
ePu
(e, ) =
ePu
ePu
2 pos(e, Pu ) =
1
2X
2i.
i=1
The last equality follows from the definition of the position of an edge on a pseudopath. Finally, a simple calculation concludes the proof of Lemma 2.10.
Now, we come to the more difficult part: obtaining an upper bound on the
negative term in (2.8).
Lemma 2.11. Let u be a level- vertex and Pu the according pseudo-path. Then
X
ePu
(e, 1)
1
1
4 + 21 .
12
3
(2.9)
Proof. Without loss of generality assume that Pu is a vertical pseudo-path that leaves
its box via the northern border.
Let N H(u) be the northern hemisphere of the box Bu of u = (i, j), i.e., N H(u) :=
{v = (i , j ) | du,v 21 , j j}. The second new definition introduces subsets of
the northern hemisphere N H(u); see Figure 2.2 for an illustration. Let S1 =
N H(v1 ) N H(v2 ) with v1 , v2 N H(u) and both v1 , v2 are level-( 1) vertices.
Further define sets Sj for j = 1, . . . , 2 recursively as follows.
Sj1 = N H(u)
N H(v)
v is
level(j1)vertex,
v (S
/ j S1 )
Geometrically, the Sj can be seen as stripes within N H(u) lying one upon the other
each time doubling their height, when incrementing j.
Notice that according to their definition the Sj do not fully partition N H(u),
but rather leave one small stripe at the bottom of Figure 2.2. Namely, we have
|Pu \
1
[
j=1
Sj | = 1.
55
S4
S3
S2
S1
u
Figure 2.2: An illustration of the stripes Sj , j = 1, . . . , 4, within the northern hemisphere N H(u) of the level-5 vertex u in G .
S
ePu
Notice that (e , 1) = 0 since the edge e was not contained in any pseudo-path
of level 1 or below. We make a simple observation.
Fact 2.12. Let u be a level- vertex and Pu its corresponding pseudo-path, which
is a vertical pseudo-path that leaves its box via the northern border. Let e be
an edge of Pu with e Pv for the pseudo-path of some level-j vertex v, j < .
Then e N H(v).
eSj Pu
(2.12)
56
because v Sj . By Fact 2.12 the edge ei Pv cannot lie between v and v, because
this would be south of v . However, since v is a level-j vertex, we have distV (ei , v)
2j1 . Now, consider the unique edge fi that lies in the pseudo-path Pv of vertex v for
which its position pos(fi , Pv ) equals distV (ei , v). Notice that Pv may be a horizontal
pseudo-path. In total, we deduce
(2.12)
ei Sj Pu
Finally, (2.10), (2.11) and (2.14) together with a simple calculation provide what
we claim in Lemma 2.11
j1
ePu
(e, 1)
1 2X
X
j=1 i=1
j1
2 pos(fi , Pv ) =
1 2X
X
2i =
j=1 i=1
1
4
4 + 21 .
12
3
Now, we go back to (2.8) and substitute the results of the two previous lemmas.
Hence,
X
eE\T
|(Se )|
k
X
k
X
=1
=1
=
=
=
X 1
4
1
1
4 +2
4 +2
4
12
3
level
vertex
1 4
4 +
6
3
(2.15)
1 k
4 k + (4k )
6
1
(N 2)2 log2 (N 2) + (N 2 )
6
1
n log2 n + (n)
12
Notice that (2.15) holds because the bounds that we derived in Lemmas 2.10 and 2.11
only depend on the level, but not on the particular level-vertices. Hence, we get the
following asymptotic lower bound.
57
Theorem 2.13. Let GN,N be the planar grid graph with n = N 2 = (2k +2)2 vertices.
For every spanning tree T of GN,N
(T )
1
n log2 n + (n).
12
Remark 2.14. In the asymptotic lower bound of Theorem 2.13 there are low-order
terms hidden in the (n). We mention two possibilities to increase these terms
as well. First, in Lemma 2.2 we only looked at either horizontal or vertical edges.
Nevertheless, an induced cut Sf on a grid always contains both types. Hence, we
actually could exploit |(Sf )| 2 pos(f, Pu,v ) + 2 for an edge f in the pseudopath Pu,v as already suggested in Remark 2.3. Second, while deriving the bound of
Theorem 2.13 we only considered edges that lie in pseudo-paths. However, it can
easily be checked that for the N N grid, with N = (2k + 2), we have |E(P)|
1/2 (N 2 N ). So, in Theorem 2.13 so far we do not consider at all the fundamental
cuts of at least 21 (N 2 N ) edges. But since all fundamental cuts have size
at least 4, we can increase our lower bound. In total, after including these three
enhancements we identify
1
28
22
42
(N 2)2 log2 (N 2) +
N2
N
6
9
9
9
as an even better lower bound.
2.3.2
bound which is 6.25n 23.5 n + 34. This will dominate the asymptotic boundin
its form in Remark 2.14not only for N 32, but rather up to N 218 . The main
work to attain this lower bound is to first establish the slightly weaker lower bound
of 6n 20 n + 22. And this work is presented mainly in the primal view on GN,N ,
with only a few times referring to predecessor edges in the rooted dual tree T .
Let C be some circuit in GN,N . We denote by diamH (C) the horizontal diameter
of C, i.e., the difference between minimum and maximum x-coordinates in Z2 of vertices in C. Similarly, we consider the vertical diameter of C, denoted by diamV (C).
It follows immediately that
|C| 2 (diamH (C) + diamV (C)).
(2.16)
For a non-tree edge e 6 T , we use diamH (e) := diamH (CT (e)) as a short hand.
58
e
diamV (e)
diamV (e)
diamH (e)
diamH (e)
(a)
(b)
Figure 2.3: In the figure, the horizontal and the vertical diameter of the circuit
induced by the non-tree edge e are illustrated. In addition, Inequality (2.16) is
motivated. For the circuit in Fig. (a), (2.16) is a proper inequality. The thin edges
can be detected as such that are counted in the left hand side of (2.16), but not in
the right hand side. On the other hand, for the circuit depicted in Fig. (b), (2.16)
holds with equality.
we
Let C be a circuit in GN,N and consider its enclosed finite region R. In C
collect all the edges in E(GN,N ) that are inside C. More precisely, these are those
edges which are incident to two faces of GN,N that both have empty intersection
with R2 \ R.
Proposition 2.15. Let GM,N be the M N planar grid, let T be a spanning tree
in it, and let and C be a simple circuit in GM,N . Then
X
\ T |.
|CT (e)| 4 |C \ T | + 6 |C
(2.17)
e(CC)\T
e(CC)\T
P
In order to show a lower bound for e(CC)\T
(diamH (e) + diamV (e)) we will define
\ T.
for all e (C C)
(2.18)
(2.19)
implying that d(e) = 2 would already satisfy (2.18). Yet, to arrive at (2.17) we must
increase d(e) beyond two. Consider the spanning tree T in the dual graph (GM,N )
that corresponds to E(GM,N ) \ T . Take F as the root of T . Consider the two
59
faces of GM,N that are incident with e. We refer to the one with the larger distance
from F in T as F (e). Note that this face F (e) is inside the circuit CT (e).
For each edge f (F (e) \ (C T {e}), denote by F (f ) 6= F (e) the other face
that f is incident with. Observe that F (f ) 6= F because this would result in a
circuit in the dual tree T . By the grid structure, each of these faces F (f ) points
into a different direction with respect to F (e), i.e., either north, east, south, or west.
Now, since f 6 T , we know that F (f ) is inside CT (e). Therefore each edge f
(F (e) \ (C T {e})) allows us to increment the lower bound on diamH + diamV ,
always satisfying (2.18). Formally, we set
d(e) := 2 + |F (e) \ (C T {e})|.
(2.20)
\ T , we now rearrange
Instead of summing up d(e) over the edges e (C C)
\ T . By the definition
the summation and count the contribution of each edge f C
of d(e) in (2.20) there are at most three edges f C \ T that contribute to the d()value of e. In turn, each such edge f is counted only for one edge e: its immediate
predecessor on the dual path from F ; formally
f (F (e) \ (C T {e})) = f
/ F (e ) \ (C T {e })
e 6= e.
So, we deduce
\ T |.
|F (e) \ (C T {e})| = |C
e(CC)\T
To summarize,
X
d(e)
(2.20)
e(CC)\T
e(CC)\T
(2 + |F (e) \ (C T {e})|)
\ T| +
2 |(C C)
=
(2.21)
(2.21)
e(CC)\T
|F (e) \ (C T {e})|
\ T | + |C
\ T|
2 |(C C)
\ T |.
2 |C \ T | + 3 |C
(2.22)
e(CC)\T
|CT (e)|
(2.16)
2 d(e)
e(CC)\T
(2.18),(2.20)
e(CC)\T
(2.22)
\ T |.
4 |C \ T | + 6 |C
Corollary 2.16. Let N 3 and let GN,N be the N N square planar grid with
n = N 2 vertices. Then for each spanning tree T E
X
(2.23)
(T ) =
|CT (e)| 6 n 20 n + 22.
eE\T
60
Proof. Simply take C as the circuit that contains precisely the edges that are incident
we apply Proposition 2.15 to C. There, we minimize the
to F . Since E = C C,
right hand side in (2.17) by maximizing |C \T |. Now consider the four vertices which
In any tree T , these must be incident to one edge
are not incident to any edge in C.
(2.24)
(T ) =
|CT (e)| 6.25 n 23.5 n + 34.
eE\T
Proof. As we assume N to be even, the graph GN,N has a unique innermost face,
which we refer to by F . Consider the pseudo-path P from F to F in the dual
graph G . Applying Lemma 2.2 and Remark 2.3 to the edges (e1 , . . . , e N ) of P
2
provides
N
(2.25)
|CT (ei )| 2i + 2, i = 1, . . . , .
2
Now, recall that in the bound in Corollary 2.16 each fundamental circuit was
(2.20)
bounded from below by at most 10, because of d(e) 5. In other words, taking
for the edges ei , i = 5, . . . , N2 the better bounds in (2.25) instead of d(e), lets us
increase the lower bound in Corollary 2.16 by
N
2
X
i=5
2(i 4).
(2.26)
2.3.3
61
For NP-hard combinatorial optimization problems often mixed-integer linear programming approaches are considered. This is also the case for the MSFCB problem
for general graphs. Different MIP formulations have been proposed by Liberti et
al. [LAM05] and Amaldi et al. [ALMM04]. In this section we qoute the MIP formulation by Amaldi et al. [ALMM04] for the MSFCB problem on general graphs.
Moreover, we make it more efficient by identifying the first classes of valid inequalities. In particular, for planar grids two of these classes are even able to cut off any
of the huge number of optimum solutions of the LP relaxation, thereby improving
the lower bound of the root node in the Branch-and-Bound tree.
Let G = (V, E) be a 2-connected graph with non-negative costs we on an edge e
E. To ensure a spanning tree T to be computed, we resort on the following characterization: |T | = |V | 1, and T is connected. We exploit the fact that T is connected, if
and only if for each non-tree edge e = {i, j} E \ T there exists a path in T between
i and j.
In their approach Amaldi et al. formulate the MSFCB problem as a multi commodity flow problem. In the following their mixed-integer linear program, see (2.27)
is developed. Amaldi et al. introduce two sets of variables: first, binary variables zij , for an edge {i, j} E, where zij = 1 if and only if e T . For the
definition of the second set of variables they introduce a directed version of G using the edge set A = {(i, j), (j, i)|{i, j} E}. For each (i, j), (j, i) A they
define wij = wji . Then, for each edge {k, } E, the variable xk
ij 0 represents the flow through arc (i, j) A from source k to sink . Observe that the
constraints (2.27b) and (2.27c) manage the flow balance between k and . Exactly
one unit of flow has to be sent. The connection between the flow an the spanning
tree is then given with Constraints (2.27d) and (2.27e). Namely, edges with strictly
positive flow have to be in the tree. Notice that since fklow is sent between all k
and with {k, } E Constraints immediately provides connectedness of the edge
set defined by z variables. Finally, the Constraint (2.27f) ensures that actually a
spanning tree is defined. In the Objective (2.27a), then, the first term collects the
costs for all paths, associated with {k, } E, and the costs of all tree chords. In
the second term the cost of the tree branches, which erroneously was counted in the
first term.
Although the MIP formulation (2.27) has been observed to behave better than
other formulations ([LAM05]), still there are some major shortcomings. First, the
number of variables and constraints is large. For instance, there are 2 m2 xvariablesin other words (N 4 ). Already with this simple observation one might
not expect too much for the solvability with, say, N 20. Still, the second drawback is even worse. The LP relaxation has several trivial optimum solutions. For
1
. This particular choice admits the x-variables to sum up
instance, take z 21 + 2N
2
to 4 (N 1) , being the optimum value of the minimum weakly fundamental cycle
basis problem on GN,N . We will provide another set of optimum solutions of the
LP relaxation in Example 2.19. Of course one can check this to be the optimum value
62
wij xk
+
ij
{k,}E (i,j)A
j(k)
k
(xk
ij xji ) = 0
xk
zij
ij
(1 2zij ) wij
(2.27a)
{k, } E
(2.27b)
{k, } E, i V \ {k, }
(2.27c)
{k, } E, {i, j} E
(2.27d)
{i,j}E
k
(xk
kj xjk ) = 1
j(i)
xk
ji
zij
zij
= n1
{i,j}E
xk
ij 0
zij {0, 1}
{k, } E, {i, j} E
(2.27e)
(2.27f)
{k, } E, (i, j) A
{i, j} E.
(2.27g)
(2.27h)
Thus, in the remainder of this chapter we provide three classes of valid inequalities: two, which are valid for general graphs and which are defined either in zvariables or in x-variables, and one class that exploits the particular structure of
grid graphs and hereby can combine x- and z-variables.
Lemma 2.18. Consider the graph G = (V, E) and an arbitrary proper subset H
of E. Denote by V (H) the vertices incident to edges in H. If V (H) ( V , then
X
X
zij |V (H)|
zij
(2.28)
{i,j}(V (H),V (H))
{i,j}H,
/
i,jV (H)
{i,j}H
63
Example 2.19. Consider the grid graph GN,N . In Figures 2.4(a) and 2.4(b) we
sketch two spanning trees T1 and T2 of GN,N . Denote the corresponding solutions of
the MIP (2.27) by (x1 , z1 ) and (x2 , z2 ), respectively.
(a) T1
(b) T2
(c) T3
Figure 2.4: The Figures 2.4(a) and 2.4(b) depict the z vectors of the MIP solutions
of two spanning trees T1 and T2 (in bold) for G9,9 . Figure 2.4(c) shows their convex
combination z3 = 12 z1 + 21 z2
Now, consider the convex combination of (x1 , z1 ) and (x2 , z2 ) which results in
the following fractional vector (x, z) := 12 (x1 , z1 ) + 21 (x2 , z2 ). Clearly, (x, z) is a
feasible solution for the LP relaxation of MIP (2.27). But observe that there exists
a vector (x , z) which is feasible for the LP, too, but in which
X X
1, if zk = 1 and
k
x ij =
(2.29)
2, otherwise.
{k,}E (i,j)A
In particular, the objective value of (x , z) equals 4 (N 1)2 . But this is just the
optimum value of the LP relaxation of MIP (2.27). As z is the convex combination of two integer feasible points, any inequality that does not make use of any
x-component, will never cut off (x , z), thus never increasing the LP value.
From the above example we conclude that for planar grids no valid inequality
having non-zero coefficients only in z-components, will ever be able to cut off one particular optimum solution of the LP relaxation, thereby never increasing the optimum
value of any so refined LP.
This is why in the sequel we investigate valid inequalities which are either defined
purely in terms of x-variables, or as a combination of x- and z-variables. But before
we do so, we prove a more general lemma that relates lengths of arbitrary cycles
in graphs to distances in spanning trees of that graph. Therefore, denote by dT (e)
the length of the unique path between the endpoints of e in a spanning tree T of an
arbitrary graph G = (V, E).
Lemma 2.20. Let G = (V, E) be a 2-connected graph with a spanning tree T and
consider a simple circuit C in G. Then,
X
dT (e) 2 (|C| 1).
(2.30)
eC
64
Proof. Let T denote an arbitrary but fixed spanning tree of G. In the following we
will prove the claim by induction over |C \ T |. Notice first that |C \ T | = 0 would
imply C T , which contradicts T being cycle-free.
Therefore, we select |C \ T | = 1 as the inductive base. In this case, the distances dT (e) in the tree are one for the |C| 1 tree edges, and |C| 1 for the unique
non-tree edge. Hence, the claim holds.
In the inductive step, take a circuit C for which |C \ T | = k 2. We will identify
two circuits C1 and C2 , each with 1 |Ci \ T | < k, i = 1, 2. Then, from (2.30) being
true for C1 and C2 we argue that the claim is true for C.
Consider two vertices u and v within C which are connected through a path P
T such that V (P )V (C) = {u, v}, but E(P )E(C) = . Such a path exists because
of |C \ T | 2, otherwise T would not be connected. Denoting by P1 and P2 the
two paths between u and v defined by C, C1 = P1 P and C2 = P2 P are simple
circuits in G.
Now, as T is cycle-free, both P1 and P2 must include at least one non-tree edge,
say eP1 and eP2 . Otherwise T would have contained a cycle. But then, C1 contains
at least one non-tree edge less than the circuit C, because it omits P2 , thus eP2 , and
the path P contains only tree-edges. The very same holds for C2 .
We may thus apply the inductive assumption to C1 and C2 . Summing up (2.30)
for these two circuits yields
X
dT (e) +
eC2
eC1
By the construction of C1 and C2 it holds that |C| = |C1 | + |C2 | 2 |C1 C2 |, and
thus
X
eC
dT (e) + 2
eC1 C2
(2.31)
eC
(2.32)
is a valid inequality for every integer feasible solution of the MIP (2.27).
65
Proof. Follows from Lemma 2.20 and the observation that for any integer feasible
solution to (2.27) it holds that
X
wa xea dT (e), for all e E.
(2.34)
aA
Note that the above corollary holds for general graphs. Yet, we are interested
most in investigating its effect on grid graphs. When considering a grid graph, a
smallest cycle has length 4. The following Lemma 2.22 generalizes Lemma 2.20 for
such cycles. Notice that (2.35) is stronger than (2.17) in the sense that (2.17) with
a LHS as in (2.35) brings a weaker bound, i.e., a smaller RHS.
Lemma 2.22. Let C be a cycle of length 4 of the grid graph GN,N . Then, a feasible
integer solution of that graph must fulfill
X X
X
xk
zk .
(2.35)
ij 18 4
{k,}C (i,j)A
{k,}C
P
Proof. A simple case distinction on the value of {k,}C zk , which is between 0
and 3, proves the claim. We refer to Fig. 2.5 for a discussion of the relevant cases.
3
1
1
1
(a)
3
3
z=3
(b)
3
1
z=2
(c)
3
3
z=2
(d)
3
9
z=1
(e)
z=0
Figure 2.5: The cases for Lemma 2.22. Tree edges are drawn in cyan, whereas for
non-tree edges brown dotted lines were used. In green we indicate the extremal cases.
That means, the green edges suggest tree edges outside the length 4 cycle C that,
together with the cyan edges, constitute a subtree that minimizes the LHS of (2.35).
For each edge of C the according flow value is depicted as well.
To summarize, in Lemma 2.22 we gave a lower bound on the sum of flow values on
a cycle of lentgh 4. Of course, one can also choose other cycles or general subgraphs
to define cuts for (2.27). However, empirical studies showed that most of them are
dominated by the cuts induced by the length 4 cycles, as presented im Lemma 2.22.
In the remainder of this section we report on experiments. For these experiments
we consider grid graphs GN,N and try to find an MSFCB using the mixed-integer linear program (2.27) which we refined with the (N 1)2 constraints of the type (2.35).
We set the weights w to 1 for all arcs of the graph. In Table 2.1 we collect information
on the MIPs and on their relaxation values.
66
Table 2.1: Some MIP stats and the relaxation values, such as the according CPU
times, are depicted. We used CPLEX version 10.1 on an Intel P4 with 3.2 GHz
and 1 GB RAM running Linux. Notice that for N = 5 the relaxation value of 68.66
brings a lower bound of in fact 70. Additionally, information on the best known
lower and upper bounds are given.
N
5
6
7
8
9
10
11
12
Best
LP relax.
LB
UB
value
time in s
72
118
176
246
328
424
528
646
72
120
184
262
356
466
592
734
68.66
110
160
222
292
374
464
566
0.5
2.1
8.1
32.8
247.6
473.6
1073.9
3097.9
MIP
#Vars
total binaries
3,240
7,260
14,196
25,200
41,616
64,980
97,020
139,656
40
60
84
112
144
180
220
264
#Constraints
#Non-zeros
4,177
9,326
18,181
32,194
53,057
82,702
123,301
177,266
17,756
40,544
80,272
143,756
238,964
375,016
562,184
811,896
Notice from Table 2.1 the huge number of variables and non-zero elements. Moreover, the time needed to solve the initial LP increases remarkably. However, observe
that the actual values for the relaxation improve on the trivial ones by a MWFCB.
Remember that using (2.27) without the additional cuts would have brought such
trivial linear relaxations with value 4 (N 1)2 . Nevertheless, when using (2.27) and
the additional cuts to optimally solve the MSFCB problem on GN,N we can only
report of a moderate performance. Whereas for N = 5 it takes less than a minute,
using CPLEX 10.1, to solve the MSFCB problem to optimality, already for N = 6
the running time of CPLEX increases to more than an hour. Of course we have
to admit that Amaldi et al. built up their MIP (2.27) for general graphs and not
particularly for grids. In the upcoming section we introduce a MIP that is especially
tailored for the MSFCB problem on planar grid graphs.
2.3.4
In this section we introduce a new mixed-integer linear program formulation for the
MSFCB problem. The MIP is specially tailored for grid graphs and we focus on
finding good relaxations. Unlike in the MIP by Amaldi et al. that we described
in the previous Section 2.3.3, we decided for a dual approach. In the following, we
explain the variables, constraints and give a detailed interpretation.
So, let GN,N be the embedded grid graph with n = N 2 vertices, see 2.2. We take
the faces of the grid as the indexing set for the variables. Therefore, we identify a
face with the coordinates of its upper left incident vertex. Define F := {(i, j) | i, j =
1, . . . , N 1} as the set of faces and let (0, 0) denote the outer face F . Now, define
67
0 otherwise,
and
(k,)
x
(i,j)
0 otherwise.
The variables x are defined for (i, j) {1, . . . N 1}2 , and (k, ) {1, . . . N 1}2
{(0, 0)} with the condition that (i, j) 6= (k, ). On the other hand, the variables x
2
2
are defined for (i, j) {1, . . . N 1} , and (k, ) {1, . . . N 1} {(0, 0)} that can
be interpreted as neighbouring faces.
The approach is to enforce the x
and the x variables to define the dual of a
spanning tree, or, more precisely, its incidence structure. Hereby, the x
determine
the actual tree whereas the x are needed to measure the length of cycles. First, we
develop some constraints that enable the interpretation of a feasible solution of the
MIP as a spanning tree. Afterwards we will motivate the objective function. Notice
that the previous will be proven in Lemma 2.23.
min
(i,j)F
HR
VT
VB
2 (dHL
(i,j) + d(i,j) + 1 + d(i,j) + d(i,j) + 1)
(k,)
(i,j)
x(k,) + x(i,j)
(0,0)
x(i,j)
(2.36b)
(i, j),
(2.36c)
(k,)
dHL
(k,) (i k) x(i,j)
(2.36d)
dHR
(k,) (k i) x(i,j)
(2.36e)
T
(j ) x(i,j)
dV(k,)
(2.36f)
B
( j) x(i,j)
dV(k,)
(2.36g)
(k,)
(k,)
(k,)
(s,t)
(s,t)
(k,)
x(i,j)
X
(k,)
x
(i,j)
(2.36h)
(2.36i)
(k,)
= 1
(k,)
= (N 1) (N 1)
x
(i,j)
(k,)
(2.36a)
x
(i,j)
(i,j), (k,)
x, x
(i, j)
(2.36j)
(2.36k)
binary
First, we enforce anti-symmetry, that is, we give the incidence structure, which
is to be established by the x
variables, a direction. For example, if face (i, j) is
68
hanging below the face (k, ) on the same dual subtree rooted by (0, 0) then the
opposite shall not be true. This is realized in (2.36b).
Then we have to ensure transitivity of the incidence structure. Namely (2.36h)
enforces that, if (i, j) is hanging below (k, ) and (k, ) is hanging below (s, t) on the
same subtree, then (i, j) is also hanging below (s, t).
Moreover, a direct incidence of faces has to imply a indirect incidence (2.36i).
And as the number of tree edges in a spanning tree obviously is fix, the number of x
2
variables set to 1 is also fix and in (2.36k) set to (N 1) . Furthermore, each face
is hanging below the outer face, which is realized in (2.36c). And last but not least
we enforce that each face has exactly one other direct predecessor on its path to the
outer face, (2.36j).
The evaluation of the length of a cycle, i.e., the objective, is motivated as follows. We introduce further variables dHL (horizontal, left), dHR (horizontal, right),
dV T (vertical, top), dV B (vertical, bottom). These variables measure the horizontal
and vertical diameter of the cycle. Namely, this works as follows. In a feasible integer
(k,)
solution, there is exactly one (k, ) with x
(i,j) = 1 for each face (i, j). Remember
that this dual tree edge encodes a primal non-tree edge. Then, for the cycle induced by that particular non-tree edge, say e, the d variables measure the diameter.
HR
VT
VB
Namely, dHL
(k,) + d(k,) + 1 equals diamH (e) and d(k,) + d(k,) + 1 equals diamV (e).
This diameter counting is done in (2.36d) to (2.36g). Obviously it holds that twice
the sum of the vertical and the horizontal diameter of a cycle is a lower bound on its
length, cf. Equation (2.16) in Section 2.3.2. However, there are two types of cycles,
those that are evaluated correctly and those whose lengths are underestimated. See
Figure 2.6 for an example.
F
(k, ) (k + 1, )
Figure 2.6: Here, an example is shown for a cycle whose length is underestimated
by the objective of the MIP (2.36). In particular, the length of the cycle induced by
the primal non-tree edge e = ((k, ), (k + 1, )) is approximated by (2.36d) to (2.36g)
VB
VT
HR
via dHL
(k,) = d(k,) = 1, d(k,) = 0 and d(k,) = 2 with a value of 12. The actual length
of the cycle, though, is 16.
Now, we prove the following on a feasible solution of the MIP.
69
Lemma 2.23. Every integer solution for the MIP corresponds to a unique spanning
tree of the grid and vice versa.
Proof. Consider a feasible solution, x
and x for (2.36). Consider the dual G of the
grid and herein the directed subgraph H defined by the edges
(k,)
(i,j)
(k,) = 1}.
{((i, j), (k, )) | x
(i,j) = 1 or x
It is sufficient to show that H is a spanning tree of G . Therefore, we show
that H is cycle free. Together with the fix number of edges in H, see (2.36k), this
implies the claim.
First, notice that because of (2.36b) there are no cycles of length 2 in H. Further,
assume that H contains a cycle of length greater than 2. Then, this cycle must be
directed, since otherwise there is an (i, j) for which (2.36j) is violated. However, along
this directed cycle one can now propagate, using (2.36i) and the transitivity (2.36h),
a contradiction to the anti-symmetry stated in (2.36b).
The statement of Lemma 2.23 together with the above mentioned fact that cycle
lengths are either calculated correctly or underestimated, we deduce that every lower
bound on the value of an optimal solution to the MIP (2.36) is a valid lower bound
for the MSFCB problem on the considered grid graph.
Table 2.2: Some MIP stats and the relaxation values as well as the according CPU
times are depicted. We used CPLEX version 10.1 on an Intel P4 with 3.2GHz
running Linux. For N = 13 the calculation in CPLEX was interrupted due to
insufficient memory. Additionally, information on the best known lower and upper
bounds are given.
N
5
6
7
8
9
10
11
12
Best
LP relax.
LB
UB
value
time in s
72
118
176
246
328
424
528
646
72
120
184
262
356
466
592
734
72
118
176
246
328
422
528
646
0.0
0.1
0.4
1.4
6.1
12.3
19.0
27.9
MIP
#Vars
total binaries
380
821
1,580
2,789
4,604
7,205
10,796
15,605
316
721
1,436
2,593
4,348
6,881
10,396
15,121
#Constraints
#Non-zeros
4,077
15,547
46,473
117,303
261,565
530,547
998,697
1,769,743
11,584
45,009
135,956
345,385
773,424
1,573,409
2,968,084
5,267,961
70
become huge. More precisely, one faces a great number of binary variables, (n2 )
even if one keeps the x variables continuous, which maintains the MIPs validity.
Moreover, the MIP contains a huge number of constraints. Namely, the transitivity constraints (2.36h) have to be defined for each triple of faces, that is (n3 ),
or, (N 6 ).
Furthermore, when the task is to optimally solve the MIP (2.36) then it turns
out that the crucial dimensions are N = 7 and N = 8. Whereas the MIP can
be optimally solved for N = 7 in approx. 70s, for N = 8 more than nine and a
half hours are needed. However, N = 8 herewith marks the largest grid for wich
the MSFCB problem is optimally solved by any MIP approach. Notice that the
obtained optimal solutions did not contain subtrees like in Figure 2.6. Therefore,
these optimal solutions of (2.36) indeed constituted optimal solutions to the MSFCB
problem. Actually, we think that these, in a sense, well-structered trees were not a
coincidence. Rather, we conjecture that the optimal value of the MIP (2.36) indeed
is the optimal value of the MSFCB problem.
Conjecture 2.24. For any optimal solution of the MIP, there is an equally valued
spanning tree for which all induced circuits are evaluated correctly by the MIP. In
other words, for any optimal solution T of the MSFCB problem on the grid GN,N
there holds that e E(GN,N ) \ T the length of its induced cycle equals twice the
sum of the cycles diameters.
Finally, a natural question that arises is, whether this MIP approach can be
generalized to a broader class of graphs. However, since the approach uses the dual
of a graph, other than planar graphs do surely not come into question. In fact, when
considering planar graphs, the encoding of a spanning tree or of its dual, respectively,
would work just as for grid graphs. Alone, counting lengths of cycles via diameters,
as in (2.36d) to (2.36g), is not possible for general planar graphs, which makes a
generalization a challenge.
2.3.5
We were able to solve the MSFCB problem on grid graphs having dimensions N 8:
for N 5 through combinatorial arguments, cf. Section 2.3.2 and especially Corollary 2.16, and for N = 6, 7, 8 through two different mixed-integer linear programming
formulations, see [ALMM04] and Sections 2.3.3 and 2.3.4. It could be observed that
the Machete-trees are optimal for dimensions N 7. However, at dimension N = 8
this structural property of an optimal solution is lost; compare the Machete trees
in Figure 2.10 and the trees in Figures 2.9(a) and 2.9(b). Nevertheless, in G8,8 the
Machete-trees are still locally optimal with respect to exchanging one tree edge with
a non-tree edge. This fact partly reflects the difficulties that a general analysis must
overcome.
Although we already solved the MSFCB problem on G8,8 , in this section we
present a combinatorial proof for a lower bound, which, together with a known
upper bound, solves the MSFCB problem for N = 8 to optimality, without using
71
(2.20)
(2.37)
and
dH (e) := 1 + |F (e) (FL (e) FR (e)) \ {C T {e}}|.
where FO (e), FL (e), FR (e) are the faces that share a border with F (e). See Figure 2.7
for an example of a horizontal edge where the F -path leaves F (e) southwards, and
remember that we may also use a face to refer to its incident edges. Of course,
we define dH (e) and dV (e) analogously for a vertical edge e. Observe that these
FO (e)
?
FL (e) F (e)
?
FR (e)
?
e
Figure 2.7: The relevant section around the non-tree edge e in G8,8 . Depending on
which of the dashed edges are in the dual tree we set d, dV and dH accordingly.
definitions ensure
dV (e) diamV (e) and dH (e) diamH (e).
Recall that in Proposition 2.15, the computation of the lower bound actually
made only use of dH and dV , which are limited to values of at most three and two,
respectively. In particular, this cannot be tight for G8,8 . Hence, in the following
definition we quantify what we were missing,
H (e) := diamH (e) dH (e) and
(2.38)
But there is another point where our analysis that led to the 6 n 20 n + 22
bound possibly was not tight. In the proof of Corollary 2.16 we used a rough bound
72
on the number of tree edges on the grids border. Later, in (2.41), we will also
improve on this.
The following series of inequalities illuminates the relation between the introduced
quantities:
X
eE\T
|Ce |
eE\T
(2.38)
eE\T
\ T|
2(H (e) + V (e)) + 4|C \ T | + 6|C
eE\T
(2.23)
eE\T
eE\T
eE\T
We may further subdivide the set C T . Let vi , i = 1, . . . , 4, be the grids four corner
vertices. We know that each of them is incident with at least one edge ei in C T .
For each of the four corner vertices vi , in (2.39) we count its corresponding edge ei
directly, thus 2|({e1 , . . . , e4 }) T | = 8. In the sequel, we only have to consider the
set C := C \ {e1 , . . . , e4 } in more detail:
X
eE\T
|Ce |
G8,8
eE\T
246 +
eE\T
(2.40)
The main work then aims to provide good lower bounds for H (e) and V (e).
However, this will not be sufficient for proving 262 to be the optimum value of the
MSFCB problem on G8,8 . In addition, we must provide lower bounds on |C T |,
and we do so by introducing values min (e) for e C \ T ,
min (e) := |{f C T | F (f ) T (e)}|,
(2.41)
where T (e) denotes all the faces that use the edge e on their path to F . Formally,
\ T we define min (e) := 0. Observe that each edge f C T is only
for e C
counted for one edge e C \ T . Hence,
X
eE\T
min (e) |C T |.
(2.42)
73
eE\T
|Ce |
(2.42)
246 +
246 +
eE\T
246 + 2
246 + 2
246 + .
min (e)
eE\T
2V (e) +
eE\T
eE\T
eE\T
(e)
(2.44)
eE\T
(2.45)
Now, by showing for an arbitrary spanning tree T that 16, we finally establish
optimality of the trees depicted in Figures 2.9(a) and 2.9(b)our ultimate goal.
In the remainder, we provide lower bounds on V (e), and on H (e) + min (e)
for particular non-tree edges e. We start by bounding from below the value H (e) +
min (e) for some e C \ T .
Lemma 2.25. Let e C \ T be a non-tree edge on the border of G8,8 . Then
H (e) + min (e) diamH (e) 1.
Proof. According to (2.38) this lemma is proven if we show that min (e) dH (e)1.
We distinguish cases according to the value of dH (e). Obviously, if dH (e1 ) 1 there
is nothing to show.
In the two cases that remain, for ease of notation we assume the edge e C \ T
to be a horizontal edge on the southern border of G8,8 . So, consider now dH (e) = 2.
W.l.o.g. the edge e FL (e)F (e) is not contained in T ; see Figure 2.8(a). Then, the
edge e FL (e) C has to be in T , because otherwise the dual tree T would contain
a circuit. Now, we distinguish two sub-cases. First, if e C we are done, because
/ C , by the
min (e) 1, c.f. the definition of min (e) in (2.41). Second, if e
definition of C the edge e must be incident to the grids corner; see Figure 2.8(b).
But then the other edge e that is incident to the very same corner has to be in the
tree, too, because otherwise we again detect a circuit in the dual tree T . Due to the
definition of C we have e C . Further notice that F (
e) T (e). Hence, min (e)
1 and we are done in the case of dH (e) = 2.
Finally, assume dH (e) = 3, cf. Figure 2.8(c). Then there exist edges e and e
with e FL (e) F (e) and e
FR (e) F (e). Hence, the edges e FL (e) C
and e FR (e) C are tree edges. By the definition of C and as we are in G8,8
we know that at least one of these two edges e and e is in C . In case both are
contained in C we observe min (e) 2 and are done. Otherwise, w.l.o.g. e
/ C .
Then, e must again be incident to a corner vertex v. Moreover, the other edge e
incident to v is in T and in C . So, we observe e , e C and F (e ), F (
e) T (e).
Therefore, min (e) 2.
74
v
e
e
F
(a) dH (e) = 2
F
(b) The
sub-case v = (1, 1)
F
(c) dH (e) = 3
)|.
and := |E(T
Notice that 1 8, because some faces share their last edge on the path to F ;
in particular the central face f5 always shares its last edge with some other face.
) = {e1 , . . . , e }. Moreover, we consider the
Further, after renaming we assume E(T
sub-trees of T that are induced by the last edge on the fi F paths. Namely
for j = 1, . . . , define
[
Tj :=
Ps .
s=1,...,9
e
s =ej
Observe that the dual sub-trees Tj are disjoint. To denote the number of fi -vertices
that are contained in Tj we define
j := |Tj {f1 , . . . , f9 }|,
for j = 1, . . . , . Further, for e E \ T recall (2.44) and let
P
eTj (e)
(Tj ) :=
j
be the relative cost with respect to the number of innermost faces that Tj contains.
Proposition 2.26. For an arbitrary spanning tree T of G
(2.45)
eE\T
(e) 16.
75
j 1, . . . , ,
P
P
then the proposition is proven, because then j=1 eT (e) 9 2 = 18.
j
Hence, in the sequel we only need to consider trees T in which there exists some
j0 1, . . . , such that
(Tj0 ) < 2.
(2.46)
Let us take a closer look at such a sub-tree Tj0 . Assume w.l.o.g. that Tj0 leaves G
via the edge e1 on its south border. In the edge sets Lk := {f E|f = {(i, k), (i +
1, k)}, i = 1, . . . , 7} for k = 1, . . . , 5 we collect the horizontal edges of the same level,
i.e., having the same vertical distance from e1 (see Figure 2.9(c) for an illustration
of L3 L5 ). To investigate Tj0 , we will consider three cases, in which we profit from
Lemma 2.25.
Case 1. {f1 , f2 , f3 } Tj0 6= . In this case, we find a sub-path Pj0 of Tj0 that starts
at e1 and which traverses Level L5 . Then, by definition of the vertical diameter,
we get diamV (e1 ) 5. Moreover let e2 be the first edge in L2 when traversing Pj0
from F and let e3 be the first edge in L3 . Then, we observe diamV (e2 ) 4
and diamV (e3 ) 3 again trivially byPdefinition. However, by definition (2.37) for all
edges e we have dV (e) 2. Thus, 3i=1 2V (ei ) 12. Now, since (Tj0 ) < 2 we
can conclude that j0 7.
But this implies immediately that diamH (e1 ) 3. Applying Lemma 2.25 to e1
yields 2 (H (e1 ) + min (e1 )) 4. In total, we have
2(H (e1 ) + min (e1 )) +
3
X
i=1
2V (ei ) 16.
(2.47)
Thus we are done with Case 1. Notice that (2.47), in fact, implies j0 = 9. Moreover,
for an arbitrary sub-tree Tj of T we showed
{f1 , f2 , f3 } Tj 6= and j < 9
(Tj ) 2.
(2.48)
76
f1 f2 f3
f4 f5 f6
f7 f8 f9
L5
L4
L3
e1
(a) A tree with
length 262.
Figure 2.9: Figures 2.9(a) and 2.9(b) show two optimal spanning trees. Notice that
the tree in Figure 2.9(b) is a representative of the trees for which all inequalities in
Case 1 hold with equality. Likewise, the tree in Figure 2.9(a) tightens the inequalities
of Cases 2 and 3.
Case 3. {f1 , f2 , f3 } Tj0 = and {f4 , f5 , f6 } Tj0 6= . Hence, Tj0 has to intersect L4 . Then, by definition of the vertical diameter, one gets diamV (e1 ) 4. Again,
denote by e2 the first edge in L2 when traversing Tj0 from F . Then, diamV (e2 ) 3.
Thus, 2V (e1 )+2V (e2 ) 6. Again, by (2.46) we use (Tj0 ) < 2 to deduce j0 4.
But this implies that diamH (e1 ) 2. Now, assume first that diamH (e1 ) = 2.
Lemma 2.25 lets us conclude that 2 (H (e1 ) + min (e1 )) 2. But this implies
6 + 2 = 8, and (Tj0 ) < 2 forces j0 to be at least five. Together with {f1 , f2 , f3 }
Tj0 = this implies diamH (e1 ) 3, a contradiction. Finally, Lemma 2.25 lets us
conclude that 2(H (e1 ) + min (e1 ) + V (e1 ) + V (e2 )) 10, and hence j0 = 6.
Similar to (2.48) we now know
{f1 , f2 , f3 } Tj = and {f4 , f5 , f6 } Tj 6= and j < 6
for an arbitrary sub-tree Tj of T .
(Tj ) 2,
(2.49)
So, what we have shown so far is the following. When we detect that Tj0 belongs
to Case 1 we are done. If this is not the case, Tj0 has to belong to Case 3. Then we
find
X
(Tj ) j .
10 +
j6=j0
Now consider a tree Tj for some j 6= j0 . Then, if Tj fulfills the condition of Case 1 we
deduce by (2.48) that (Tj ) 2. Analogously, if Tj belongs to Case 3 we use (2.49)
to seeP
(Tj ) 2. For the remaining Case 2, we also deduce (Tj ) 2. So, finally,
since j6=j0 j = 3 we have
10 + 2 3 = 16.
This concludes the proof of Proposition 2.26.
Now, the major work is done to obtain our final result:
77
Upper Bounds
In this section we report on upper bounds on the MSFCB problem on grid graphs.
The previously best known spanning trees are from Alon et al. [AKPW95]. They give
trees with an induced cycle basis length of O(n log n) with n beeing the number of
vertices of the grid GN,N , i.e., n = N 2 . They also showed that this is asymptotically
best possible. However, we consider upper bounds of the form c n log2 n + o(n log n)
with a constant c. In Section 2.4.1 we develop spanning trees that reach an upper
bound of the above form with c = 0.979. This improves on the constant that reach
the trees of Alon et al. by a factor of more than four third.
Not surprisingly, such asymptotic bounds only work out at very large dimensions
since they make extensive use of recursive subtree structures. So, at the end of
Section 2.4.1 we propose adapted spanning trees that we show to be well-suited for
dimensions N 100 in the experiments presented in Section 2.5.
However, we start this upper bounds section by taking a look at the really small
dimensions. For example, if one considers spanning trees for N = 3, 4, 5 an alleged
structure is detectable. Even more, if ones checks such trees for dimensions N =
6, 7 the conjecture seems to approve. For N = 3, . . . , 7 this specially structured
4 2
N 3
2
i
XX
i=1 j=1
N 1
2
(2j + 2) +
X
i=1
3
13
1
(2i + 2) = N 3 + 2N 2 N + 2 = n 2 .
3
3
78
(a)
(b)
(c)
To assess the value of f (N ), in accordance with Fig. 2.11(b) we group the summation
into five blocks: four on the vertical part of the middle cross (top-down), plus its
horizontal part:
N
N
4
f (N ) =
1
1
1
4
4
X
X
X
(2i + 2) +
(3N 4 + 2i) +
(3N 2 + 2i) +
i=1
N
4
i=0
X
i=0
(4(N 1) + 2i) + 2
i=0
N
2
X
i=1
(2i + 2) =
13 2
N 2N 6.
4
79
Although Alon, Karp, Peleg, and West ([AKPW95]) think of their trees as being
essentially optimal, we are able to construct trees with an asymptotic coefficient
for the n log2 n term being strictly smaller than one. Namely, we provide spanning
trees that induce strictly fundamental bases of size not more than 0.979 n log2 n +
o(n log2 n). These spanning trees will be defined for large dimensions. However,
at the end of this section we present similar trees which empirically perform very
well already in small dimensions. Then, we make use of this in the experiments
Section 2.5.
Before we describe how we construct the asymptotically good spanning trees, with
the next paragraph we motivate how a class of recursively defined trees looks like that,
in fact, accommodates both of the above mentioned goals. These spanning trees are
the union of spanning trees in rectangular subgraphs of GN,N , their building blocks.
The trees differ in how their rectangular subgraphsall respecting some arbitrary
but fixed aspect ratio 1partition the faces of GN,N . Hence, it remains to
specify how to construct a spanning tree subject to a given parameter for some
N
grid GM,N having aspect ratio max{ M
N , M } . This is done recursively. Assume
w.l.o.g. that M N . At the top-level of the recursion, we add to T (GM,N ) the
edges of the two longer borders of GM,N (here the horizontal ones), plus of one of
its two other borders (cf. Figure 2.12). For the recursion, we partition the faces of
GM,N into almost equally-sized rectangular subgraphs of aspect ratio again being
close to ; only the faces of one horizontal path in (GM,N ) , located almost in the
middle of its two horizontal borders, are not contained in any of these rectangular
subgraphs.
80
...
... ...
...
sub-sub-block
sub-block
Figure 2.12: The shape of a block (left) and with a sketched interior recursively filled
with smaller blocks (right), always keeping the aspect ratio.
These trees are related to other families of trees as follows. In GN,N , choosing
N2 : 1 there exists a partition of the grid such that we end with Machetetrees ([Bok03] and [BKW03], cf. Figure 2.10). Moreover, an aspect ratio of = 1 : 1
yields trees which can be obtained alternatively by a construction that is much similar
to the one for TAKP W .
According to the requirementasymptotical or empirical qualityone can consider trees with a block structure either having an aspect ratio of approximately 3 : 1
or an aspect ratio of 2 : 1, respectively. In addition, the trees differ in how the blocks
are actually used to define a tree. Whereas on large grids it is sufficient to cover the
grid with three (almost) equally-sized 3 : 1 blocks, for small dimensions the grids are
tiled with many 2 : 1 blocks of many different sizes.
To achieve a good asymptotical upper bound we decided to construct trees out
of the above described blocks with an aspect ratio of 3 : 1. Unfortunately, it turns
out to be tricky to subdivide or tile a square grid of arbitrary dimension with these
particular blocks. Thus, we construct our trees bottom-up like. That means we take
an atomic block of size 6 14 and arrange 32 copies of such a block to a new one
having size 80 14. This procedure is then iterated providing spanning trees for
dimensions
15
1248
32k/2 +
496
31
419
30
32k/2 +
496
31
(2.52)
with k chosen integral and even. Finally, three copies of such a tree can be put
onto each other and cover the entire square grid. Now, a detailed description of the
construction of the tree and a precise analysis of it follow.
81
y2i = 2 y2i1 + 1
(2.53)
for trees Tk with k = 2i even. For odd k = 2i + 1 the tree Tk has dimension
x2i+1 = 2 x2i + 1
y2i+1 = 16y2i .
(2.54)
In the following we will only consider the spanning trees Tk for k even. Notice
that the Tk with k even, always have their exit on the left vertical border of the grid.
Now, simple calculations transform (2.53) and the start values x2 = 80 and y2 = 27,
respectively, into the explicit sequence of the side-lengths of Tk for even k:
xk =
78
16
32k/2
31
31
yk =
419
1
32k/2 .
496
31
(2.55)
If we now take a closer look at Tk , k even, we see that the ratio of its lengths
is almost 3 1. In fact, the exact ratio xk to yk is always greater than 2.96 and
converges to 1248
419 2.978.
Hence, if we take three copies of Tk and put them one upon another, then the
resulting spanning tree, let us denote it by Tk3 , covers a grid of dimension
1248
15
32k/2 +
496
31
1257
30
32k/2 +
496
31
82
We now claim that three times the size of Tk is an upper bound on the size of Tk3
restricted to the square grid G with dimension
1248
15
15
1248
k/2
k/2
32 +
32 +
.
496
31
496
31
So, how do we restricted Tk3 to a square of the above size? Let us consider the
boundary line L of G that, in a sense, cuts through the down-most copy Tk of Tk3 ,
cf. Figure 2.14. This Tk consists of several subtrees Tk1 , Tk3 , . . . , T1 . Those odd
subtrees can have their exit pointing downwards, Tj , or upwards,denoted by Tj .
If for a j = 1, 3, . . . , k 3, k 1 the boundary line L cuts through a subtree Tj
we leave this part of tree unchanged and simply cut away what overhangs L. In the
other case where the boundary line L cuts through a subtree Tj we cut away the
overhanging parts as well, butsince we loose connectivitywe add an edge to Tj
exactly where formerly the exit had been. If we do so for all j = 1, 3, . . . , k3, k1 we
finally come up with a tree, let us denote it with Tk whose chords induce cycles with
lengths not greater than they had been before, i.e., within this down-most copy Tk
of Tk3 .
Now, we continue with the analysis of Tk .
. . . 16 copies . . .
Tk2
f2
f2
f2
Tk1
f1
Tk1
...
...
Figure 2.14: A schematic illustration of the tree Tk for an even k + . Due to the
construction rules Tk consists of 322 = 1024 copies of Tk2 and different slots, i.e.
one main slot f1 , dark-gray, and 32 subslots f2 , light-gray.
Analysis of the tree. So, for a k + and k even, consider the spanning tree
Tk of the square grid with dimensions
1248
1248
15
15
k/2
k/2
32 +
32 +
.
496
31
496
31
83
We are interested in an upper bound on the strictly fundamental cycle basis induced
by Tk . As argued above we have
(Tk ) 3 (Tk ).
(2.56)
In the following we develop a recursive formula for (Tk ). Because of the trees
special construction the following recursive formula holds,
(Tk ) = 1024 (Tk2 ) + f (Tk ),
(2.57)
where f (Tk ) denotes the size of the fundamental cycles induced by edges that do
not lie entirely within a copy of the smaller Tk2 tree. We call those areas slots.
Then, f (Tk ) can be canonically subdivided into one main-slot and several sub-slots,
cf. Figure 2.14. Obviously,
f (Tk ) = f1 (Tk ) + 32 f2 (Tk )
(2.58)
holds. Then, with the help of the sequences defining the lengths of the trees (Equations (2.53) and (2.54)) we straight-forward express f1 and f2 as
1
x +1
32 k
f1 (Tk )
2i +
i=1
15
X
j=1
x +1
32 k
X
i=1
x +1
k
32X
2j
(2yk + xk + 2i) +
(2xk + 2yk + 2i),
16
i=1
(2.59)
and
1
y
+1
32 k1
f2 (Tk )
2i +
i=1
15
X
j=1
y
+1
32 k1
i=1
1
y
+1
32 k1
2j
(2xk1 + yk1 + 2i)
16
(2.60)
(2.61)
i=1
respectively. Further, plugging (2.59) and (2.60) into (2.58) and then (2.58) into (2.57)
we yield the recursion:
(Tk ) 1024 (Tk2 ) +
20,323,353
32k + o(32k ).
984,064
(2.62)
Here, we omit the value for the recursion start T2 because it is of no importance for
the coefficient of the n log2 n term.
After resolving (2.62) and applying the result to (2.56) one gets
(Tk )
60,970,059
32k k + o(32k k).
1,968,128
78
15
n=
32k/2 + ,
31
31
84
6, 774, 451
n log2 n + o(n log2 n).
6, 922, 240
Remember that thereby the previously best asymptotical upper bound by Alon
et al. ([AKPW95]) is enhanced by a factor of more than four third.
Now, as mentioned before the description of the asymptotical good spanning
trees, we consider similar trees that induce short SFCB already for small dimension. The 3 : 1block structured trees, as described in the above paragraph are not
perfectly suited for smaller dimensions. As shown, 3 : 1 is asymptotically a very
good aspect ratio. Yet, it is not possible to decompose an arbitrary square grid into
3 : 1blocks without losing much of their advantage because of rounding errors.
Therefore, for small grids, we chose a different block-structured graph. This time
we use an aspect ratio of 2 : 1. In contrast to the above, the 2 : 1blocks, do not
really cover, but rather tile the square grid. The tiling procedure roughly goes as
follows:
At first, two opposite 2 : 1blocks are put in the middle of the grid. See for
example the two blocks marked with A, having side lengths 8 15 in Figure 2.15.
Next, horizontal 2 : 1blocks (marked with B) are added centrally aside such that
rectangular subgrids in the four corners remain. In those corners (marked C) we
always direct the next block such that its depth can be chosen as small as possible,
while its aspect ratio should stay as close as possible to the target ratio 2 : 1. During
this procedure we do not need to pay attention to any rounding inaccuracies. In
Figure 2.15 an example 2 : 1block structured tree for dimension N = 31 is shown.
The empirical quality of the so defined trees for small grids will be evaluated in
the next section.
2.5
Experiments
In this section we compare different spanning trees with respect to the length of
the strictly fundamental cycle basis they induce. The experiments conducted in this
section are intended to provide benchmark results considering both upper and lower
bounds for dimensions N = 5, . . . , 100.
85
2.5 Experiments
C
B
B
B
C
A
A
C
B
B
B
C
Figure 2.15: Notice the parquet-like structure of the tree with tiles having heightwidth ratio of 2 with small errors due to roundings. Inside, the blocks themselves
are recursively filled with smaller blocks still maintaining the 2 : 1 ratio.
(a)
(b)
Figure 2.16: In Figure (a) an edge swap is illustrated: the red edge induces a fundamental cut out of which the green edge is taken to anew constitute a spanning
tree. In this example, the edge swap decreases the size of the induced strictly fundamental basis. However, one may get stuck when using edge swaps to improve the
basis. This is shown in (b). There, although the tree is locally optimal, the blue
highlighted subtree causes the non global optimality.
is not exact. That means, see Figure 2.16(b) for example, there are local optimum
trees, which are not globally optimal.
Amaldi et al. ([ALMM04]) reported the performance of several strategies for
searching the neighborhood of a spanning tree. In what they denote by local search (LS),
86
the entire neighborhood is examined and they move to the tree with the best improvement. In a second deterministic strategy (ES, for local search with edge sampling)
only a restricted number of neighbors are tested.
To prevent LS to terminate too early in a too bad local optimum, Amaldi et
al. ([ALMM04]) run metaheuristics such as variable-neighborhood search (VNS) and
a tabu search (TS) on top of LS. In any of their computations, an adapted version
of the tree-growing heuristic of [Pat69] is used as the initial solution.
In our computations, we use the 2 : 1block-structured tree as initial solution.
In contrast to (LS) we do not examine the entire neighborhood for improvement.
Instead, whenever we identify a neighbor that improves the current solution, we
greedily move to that neighbor. Of course, this method depends on the order in
which the edges in the tree are checked. Empirical studies showed, however, that the
influence of the edge-order is neglectable. For our computational studies we chose
a random order of edges and ran our greedy-like approachdenoted by (GS)ten
times, considering the best value of the length of the cycle basis and the according
running time of (GS). We skip average values, because we see the goal of the study in
giving benchmark results. Examining the quality of the heuristic is only a secondary
goal. Among the ten sample runs the lengths of the cycle bases vary by less than 1%
only, anyway.
In Table 2.3 we compare the constructive heuristics, i.e., those that build up a
tree without doing any subsequent local improvements. Moreover, we complement
these values with information on lower bounds obtained by Corollary 2.16, for odd
dimension, and by Theorem. 2.17 for dimensions N 10 and N even, as well as with
values of a minimum cycle basis.
The latter were also used in the recent study of Amaldi et al. [ALMM04]. Note
further, that N = 130 is the dimension closest to 100 for which our asymptotic lower
bound, see Remark 2.14 is defined exactly. We mention that for this value of N
the bound that we derived in Theorem 2.17 is by more than 40% stronger than the
asymptotic bound.
In our tables the italic numbers highlight the best known upper and lower bounds.
For N = 5, these coincide and we mark this in boldface. Observe that for any
dimension N 10, the new trees that we proposed at the end of Section 2.4.1 yield
smaller SFCB values than any of the other constructive heuristics.
In Table 2.4 we compare the different local-search-type heuristics. For our greedy
search (GS) we used a 3.2GHz Intel P4 computer (A1), running Linux and using
c
LEDA
. Amaldi et al. used for their local search heuristics (LS) and (ES) also an
Intel P4 computer running Linux, but with 2.66GHz (A2). Accordingly, the times
stated in Table 2.4 refer to the particular architecture. The values for the metaheuristics (TS) and (VNS)also quoted from [ALMM04]each refer to 10 minute
runs on the A2 environment. Much similar as in the purely constructive context, our
new solutions improve the best known upper bounds for all dimensions N 20.
As already mentioned before we ran our local search (GS) with a random order
of the edges. In Table 2.4 the first two columns present the value for the best run
out of ten samples, and the according running time, respectively.
87
Table 2.3: Comparison of the cost of some selected trees, i.e., the length of the according strictly fundamental cycle bases. The rightmost column presents the previously
best lower bound for small dimensions, obtained just by 4(N 1)2 . The penultimate
column now states the consistently better lower bounds due to Corollary 2.16 and
Theorem 2.17.
N
2 : 1
5
10
15
20
25
30
35
40
45
50
55
60
70
80
90
100
76
468
1 300
2 550
4 368
6 656
9 592
13 162
17 236
21 920
27 356
33 406
47 300
63 964
83 412
106 090
1
3
5
8
11
16
21
27
35
42
59
80
108
137
78
524
554
030
410
408
694
078
784
912
124
790
244
678
012
390
1
3
6
10
16
24
34
46
61
78
123
183
258
352
72
492
512
382
352
672
592
362
232
452
272
942
832
122
812
902
1
3
6
10
16
24
34
46
61
78
72
492
512
382
352
672
592
362
232
452
272
942
1
3
6
11
16
28
35
48
62
92
78
518
588
636
452
638
776
100
744
254
026
978
1
2
3
4
6
9
11
14
17
21
29
38
48
60
72
424
072
064
272
954
672
094
272
484
072
124
014
154
544
184
1
2
3
4
6
7
9
11
13
19
24
31
39
64
324
784
444
304
364
624
084
744
604
664
924
044
964
684
204
However, it has to be mentioned that only for dimensions N {60, 80, 90, 100}
the start tree had not already been locally optimal.
2.6
In this chapter we investigated the Minimum Strictly Fundamental Cycle Bases (MSFCB) problem on grid graphs.
First, we gave a new proof of an (n log n) asymptotical bound on the size of
an optimal basis that uses dual graphs and a new concept of pseudo-paths. A
detailed analysis revealed that, w.r.t. the factor before the n log2 n, we enhanced
the previously best lower bound by a factor of more than 245. Then we switched to
small dimensions where we combinatorially developed a lower bound which, although
asymptotically not optimal, turns out to be the best lower bound for dimensions up
to N = 218 . Moreover we considered two different MIP formulations and established
new cuts for them. Then, for dimension N = 8, we identified a tight lower bound
hereby for the first time proving 262 to be the optimal value of the MSFCB problem
for G8,8 . As for the asymptotical upper bounds, we developed a family of spanning
88
Table 2.4: An overview of the quality of five local search approaches. Missing
values are marked with an and running times are measured in mm:ss. The
columns (LS)(TS) are cited from [ALMM04].
N
5
10
15
20
25
30
35
40
45
50
55
60
70
80
90
100
(GS)
cost
time
72
468
1 300
2 550
4 368
6 656
9 592
13 162
17 236
21 920
27 340
33 374
47 300
63 810
83 222
105 766
00:00
00:00
00:00
00:00
00:00
00:01
00:02
00:07
00:06
00:09
00:31
01:01
00:44
07:24
07:48
14:01
1
2
4
6
10
13
18
23
(LS)
cost
time
(ES)
cost
time
(VNS)
cost
(TS)
cost
72
474
318
608
592
956
012
548
100
026
74
524
430
186
152
488
662
924
602
274
72
466
1 280
2 572
4 464
6 900
9 982
13 524
18 100
23 026
72
466
1276
2590
4430
6882
9964
13534
18100
23552
00:00
00:00
00:00
00:03
00:16
00:47
02:19
06:34
14:22
31:04
1
3
5
8
11
15
22
33
00:00
00:00
00:00
00:00
00:02
00:03
00:08
00:26
01:00
01:10
trees improving on the previosly best known ones by a factor of more than 43 . And,
last but not least, we conducted some experiments that establish benchmark results.
Next, we list some open questions. A first naturally question is, whether the gap
between the lower bound and the upper bound on an MSFCB on a grid can be further
reduced. Then, concerning mixed-integer linear programs for the MSFCB problem
it remains to build specially tailored formulations for planar graphs. Moreover, the
complexity status for the MSFCB problem remains open for grid graphs and planar
graphs.
We conclude this chapter with depicting spanning trees, see Figs. 2.17 and 2.18,
which constitute best known upper bounds for dimensions N = 9, . . . , 20. With the
exception of the tree for N = 20, which was provided by [DK08], these trees were
found by hand, rather than by a certain heuristic.
89
(a) N = 9:
356
(b) N = 10:
466
(c) N = 11:
592
(d) N = 12:
734
(e) N = 13:
894
(f) N = 14:
1072
90
(a) N = 15:
1268 (1276)
(b) N = 16:
1486
(c) N = 17:
1720
(d) N = 18:
1972
(e) N = 19:
2244
(f) N = 20:
2534
3
CLASSIFICATION OF TREE SPANNER
PROBLEMS
In this chapter we deal with tree spanner problems. Tree spanner problems have
important applications in network design, e.g. in the telecommunications industry.
Mathematically, there have been considered quite a number of max-stretch tree spanner problems and of average stretch tree spanner problems. We propose a unified
notation for 20 tree spanner problems, which we investigate for graphs with general
positive weights, with metric weights, and with unit weights. This covers several
prominent problems of combinatorial optimization. Having this notation at hand,
we can clearly identify which problems coincide. In the case of unweighted graphs,
the formally 20 problems collapse to only five different problems. Moreover, our
systematic notation for tree spanner problems enables us to identify a tree spanner
problem whose complexity status has not been solved so far. We are able to provide an NP-hardness proof. Furthermore, due to our new notation of tree spanner
problems, we are able to detect that an inapproximability result that is due to Galbiati [Gal01, Gal03] in fact applies to the classical max-stretch tree spanner problem.
This chapter is based on [LW07].
3.1
Introduction
We consider a weighted connected undirected graph (G, w), where G = (V, E). We
assume the edge weights to be positive integers, occasionally after scaling. Let T be
a spanning tree of G. Depending on the context, we think of T either as a subset of
the edges of G, or as a subgraph of G. For a spanning subgraph H of G and u, v V
we denote by
dH (u, v)
the length of a shortest (u, v)-path in H.
91
92
In [CC95] the t-tree spanner problem has been introduced as follows: Decide
whether there exists a t-tree spanner, i.e., a spanning tree T of G such that
dT (u, v)
t,
dG (u, v)
(u, v) V V := V V \ {(v, v) | v V }.
(3.1)
{u,v}E\T
dT (u, v)
.
dG (u, v)
(3.2)
Since for unweighted graphs there holds dG (u, v) = 1 for all {u, v} E, it is a simple
observation ([AKPW95]) that this MAST problem turns out to be nothing but a
special case of the Minimum Strictly Fundamental Cycle Basis (MSFCB)
problem as it has been considered for instance in [DKP82]: Find a spanning tree T
that minimizes
X
dT (u, v) + w(e).
(3.3)
e={u,v}E\T
dT (u, v).
(3.4)
(u,v)V V
The very same problem has also been referred to as the Minimum Routing Cost
Spanning Tree (MRCST) problem ([WLB+ 99, GA03]). In the special case of an
unweighted graph, Johnson et al. ([JLK78]) call it the Simple Network Design
problem. Alternatively, when considering complete graphs, Hu ([Hu74]) introduced it
as the Optimum Distance Spanning Tree problem. In [DDGS03], the Minimum
Average Distance (MAD) spanning tree problem is consideredbut setting the
vertex weights in that model to one, this is another variant of the STPLST problem.
Notice that also additive tree spanner problems attracted quite a number of researchers (e.g. [KLM+ 03]). However, the work presented in this chapter is restricted
to max-stretch and average stretch tree spanner problems.
93
In the following Section 3.2 we propose a unified notation for the large variety of
tree spanner problems. Subject to this notation we identify which problems coincide.
More specifically, we consider two problems P and Q to coincide, if every spanning
tree TP that is optimum for P constitutes an optimum solution for Q, and vice-versa.
We use the notation P Q as a short-hand. Notice that we have to choose this
very discriminative equivalence relation. Otherwise, if we allowed for general polynomial transformations, one could no more distinguish between any two NP-complete
problems. We provide coincidences for both maximum stretch tree spanners (Section 3.3.1) and average stretch tree spanners (Section 3.4.1), for the cases of graphs
with general weights, with metric weights, and with unit weights. We complement
our analysis by providing example graphs showing that there are no further coincidences. All the examples consist of fairly small simple 2-vertex connected planar
graphs.
Consider the very rich world of (in-) approximability results for tree spanner
problems, occasionally for special classes of graphs. We expect that having at hand
a clear map of the relationships between the various tree spanner problems, a certain
cross-fertilization between the different perspectives on much similar structures will
occur. In Section 3.6.2 we make the first step into this direction. In the context
of tree spanners, as recently as 2004 the
best known inapproximability factor of the
1+ 5
MMST problem has been cited as 2 ([EP04]), being due to [PR99]. During this
chapter we show that in the case of unweighted graphs the MMST problem coincides
with the Min-Max Strictly Fundamental Cycle Basis (MMSFCB) problem,
as it has been stated in [Gal01, Gal03]. There, an inapproximability factor of 2
has been achieved already in 2001 ([Gal01]). Hence, this applies immediately to the
MMST problem as well. Moreover, in the family of tree spanner problems we identify
the only problem whose complexity status has not been identified before. We provide
an NP-hardness result for it.
3.2
There are three major criteria in which tree spanner problems may differ: First,
either the maximum stretch or the average stretch is to be determined. Second, this
objective may be computed with respect to different sets of pairs of vertices, e.g. for
(u, v) V V or only for {u, v} E \ T . Third, there have been considered various
T (u,v)
terms for the objective, e.g. ddG
(u,v) or dT (u, v) + w(e).
In the remainder, we refer to a tree spanner problem P through a triple
(goal, domain, term) .
We consider the following family of tree spanner problems:
goal
The goal is either the maximum stretch or the average stretch.
domain
The domain is either {u, v} E \ T , {u, v} E, or (u, v) V V .
94
term
(u,v)
T (u,v)
, or ddG
The term may be one of dT (u, v) + w(e), dT (u, v), dTw(e)
(u,v) .
(u,v)
Notice first that we do not consider , V V, dTw(e)
and (, V V, dT (u, v)+
w(e)), because w(e) is not properly defined for (u, v) V V \ E. Second, it
could appear somehow strange to count the weight of tree edges twice in the two
tree spanner problems (, E, dT (u, v) + w(e)). However, this is consistent with the
UNTS. Moreover, this does not cause any degeneracies, because in the next two
sections we exhibit that there is always some other tree spanner problem, which
coincides with (, E, dT (u, v) + w(e)). Third, observe that for a given graph, |E|
and |V V | are constant, and |E \ T | is independent
of the tree T . Hence, we prefer
P
to represent the goal average with the
symbol.
We provide a first idea of the wide range of these tree spanner problems by locating several well-known problems of combinatorial optimization within the UNTS:
T (u,v)
max, V V, ddG
(u,v) is the MMST problem ([CC95]),
(max, V V, dT (u, v)) is the Minimum Diameter Spanning Tree (MDST)
problem ([HL02]),
P
( , V V, dT (u, v)) is the STPLST problem (or MRCST problem, [JLK78]),
and
P
( , E \ T, dT (u, v) + w(e)) is the MSFCB problem ([DKP82]).
We will establish that among the remaining 16 problems there is only one single
problem which does not coincide with one of these four prominent problems in the
case of unweighted graphs. Since its complexity status has not been identified before,
we provide an NP-hardness proof for it. However, in the case of weighted graphs
there is a much larger variety of problems, in particular in the context of average
the same techniques are applied to
stretch
tree spanners.
in [EEST05]
For instance,
P
P
dT (u,v)
dT (u,v)
and
, E, dG (u,v) . Nevertheless, in general these problems
both
, E, w(e)
do not coincide.
In Figure 3.1 we summarize all the coincidences that exist between tree spanner
problems, and which we are going to develop in the remainder of this chapter.
Namely, in Sect. 3.3 we deal with max-stretch problems whereas in Section 3.4
average-stretch problems are considered. We organize the sections by subdividing them into three parts where we distinguish general, metric, and unit weights.
However, we show that there is a bridge between unweighted and integer-weighted
tree spanner problems. Here, we aim at identifying a weighted instance (G, w)
immediately with the unweighted instance G that results from replacing every
edge e = {u, v} with weight w(e) with a uv-path Pe having w(e) edges. Observe
that every spanning tree of G has to contain at least w(e) 1 edges of Pe . Now,
consider the term dT (u, v) + w(e). Let T be some spanning tree of G. We construct
a spanning tree T of G such that if e T then Pe T . This yields
dT (u, v) + w(e) = dT (u , v ) + 1,
e = {u, v} E \ T, {u , v } = Pe \ T .
(3.5)
95
Hence, for the domain E \T in conjunction with the term dT (u, v)+w(e) an optimum
solution to a weighted tree spanner problem is obtained by a kind of projection from
an optimum solution to the corresponding unweighted problem, and vice-versa.
Proposition 3.1. Let goal be a fixed optimization goal. Then, the weighted version
and the unweighted version of (goal , E \ T, dT (u, v) + w(e)) coincide.
3.3
We start our tour through the zoo of tree spanner problems with maximum stretch
tree spanner problems. We first collect the pairs of problems which coincide, where
we distinguish between general weights, metric weights, and unit weights. Then, we
examine example graphs showing that there are no further coincidences.
3.3.1
Coincidences
It is an elementary observation that if two tree spanner problems coincide even for
general weights, in particular they also coincide for metric weights. Moreover, if
two problems coincide for metric weights, they immediately coincide for unweighted
graphs, too. Hence, to present the coincidences between maximum stretch tree spanner problems, we proceed from the most general weight functions to the most specialized weight function.
General Weights. In the case of general weights, there are five families of coincident maximum stretch tree spanner problems.
Proposition 3.2. The following two maximum stretch problems coincide: (max, E \
T, dT (u, v) + w(e)) and (max, E, dT (u, v) + w(e)).
Proof. Assume for contradiction there was a weighted graph (G, w) such that a
spanning tree T that is optimum with respect to (max, E, dT (u, v) + w(e)) attains
its maximum exclusively on a tree edge e T . Then,
f = {u, v} E \ T : dT (u, v) + w(f ) < 2w(e).
(3.6)
Consider any edge f E \ T whose fundamental circuit C contains the edge e. Such
an edge exists because G is 2-vertex connected. The total weight of the circuit C is
precisely w(C) = dT (u, v) + w(f ), and the weight of C \ {e} is dT (u, v) + w(f ) w(e).
By (3.6) there holds
dT (u, v) + w(f ) w(e) < w(e).
(3.7)
Now, consider the spanning tree T = T {f }\{e}. Any fundamental circuit (different from C) with respect to T that contained the edge e is replaced with a subpath
of C \ {e}. As we only consider positive edge weights, by (3.7) the new fundamental
circuit is strictly shorter than the initial one. Hence, the spanning tree T has not
been optimum with respect to (max, E, dT (u, v) + w(e)).
96
dT (u,v)
w(e)
dT (u,v) dT (u,v)+w(e)
E\T
11111111
00000000
00000000
11111111
00000000
11111111
00000000
11111111
E\T
unweighted
11111111
00000000
00000000V V
11111111
00000000
11111111
00000000
11111111
E
max
P
11111111
00000000
00000000
11111111
00000000
11111111
00000000
11111111
11111111
00000000
00000000
11111111
00000000
11111111
00000000
11111111
E\T
11111111
00000000
00000000
11111111
00000000
11111111
00000000
11111111
metric
11111111
00000000
00000000
11111111
V V
00000000
11111111
00000000
11111111
E
max
P
E\T
E
V V
11111111
00000000
00000000
11111111
00000000
11111111
00000000
11111111
dT (u,v)
dG (u,v)
dT (u,v)
w(e)
dT
11111111
00000000
00000000
11111111
00000000
11111111
00000000
11111111
d (u,v)+w(e)
(u,v)
T
E\T
E
V V
weighted
MMST
11111111
00000000
00000000
11111111
MDST
00000000
11111111
00000000
11111111
11111111
00000000
00000000
11111111
00000000
11111111
00000000
11111111
MSFCB
max
P
11111111
00000000
00000000
11111111
00000000
11111111
00000000
STPLST 11111111
00000000
11111111
00000000
11111111
00000000 d (u,v)
11111111
00000000
11111111
d (u,v)+w(e)
dT (u,v)
dG (u,v)
dT (u,v)
w(e)
Proposition 3.3. The two problems (max, E \ T, dT (u, v)) and (max, E, dT (u, v))
coincide.
Proof. Let T be an arbitrary spanning tree of (G, w). These two problems could
97
only differ, if the maximum in (max, E, dT (u, v)) is attained exclusively by a treeedge e = {u, v} T . But in this case, dT (u, v) = w(e). As we only consider 2-vertex
connected graphs, there exists a circuit C through e. The tree T cannot contain all
the edges of C. Hence, as we assume the weight function w to be positive, there
exists a non-tree edge e = {u , v } C \ T such that dT (u, v) dT (u , v ).
(u,v)
In the sequel we establish that the following fiverecall that max, V V, dTw(e)
(u,v)
is not properly definedmaximum stretch problems coincide: max, , dTw(e)
and
T (u,v)
max, , ddG
(u,v) . In fact, most of the work was done by Cai and Corneil ([CC95]):
dT (u,v)
dG (u,v) ,
dT (u,v)
dG (u,v)
(u,v)
path in G. Thus
1 for all (u, v) V V . When considering the term dTw(e)
over the domain E, for every tree edge e T this edge constitutes the unique uv-path
(u,v)
in T . In particular, dTw(e)
= 1 for all e = {u, v} T E.
Proposition 3.6.
If a weighted graph
(G, w) admits a tree spanner T such that
dT (u,v)
t < 1 subject to max, E \ T, w(e) , then T is unique optimum for all five prob
(u,v)
T (u,v)
lems max, , dTw(e)
and max, , ddG
(u,v) .
max, , ddG
(u,v) have stretch factor t > 1.
98
CT (f ),
f F
where we consider the symmetric difference. Due to the special structure of cycle
bases that are associated with spanning trees, we know that F CT (e) \ T , in
fact F = F ([Ber62]). In particular, as by definition the edge e is contained in CT (e),
e has to appear in at least one fundamental circuit CT (f ) that is induced by an
edge f = {u , v } F .
Corollary
3.7.The following
five maximum
stretch tree spanner problems coincide:
dT (u,v)
dT (u,v)
and max, , dG (u,v) .
max, , w(e)
dT (u,v)
Proof. Consider an optimum solution T with respect to max, E \ T, w(e) . In
the case of a stretch factor t 1 we are done immediately by applying Theorem 3.4.
Otherwise, i.e., if t < 1, Proposition 3.6 ensures optimality and uniqueness of T
subject to all five optimization problems that we consider here.
In particular, all the maximum stretch tree spanner problems that involve fractions coincide.
99
Metric Weights. For maximum stretch tree spanner problems, there are no coincidences in the case of metric weights, which do not apply already to the general
case.
Unit Weights. For an arbitrary tree T of an unweighted connected graph G (or
having weights w = 1) with n vertices and m edges, there holds
max
e={u,v}E
max
e={u,v}E\T
max
e={u,v}E
max
{dT (u, v) + 1}
e={u,v}E\T
{dT (u, v) + 1, 2} .
dT (u,v)
w(e) .
(3.8)
(3.9)
(3.10)
dT (u,v)
dG (u,v) .
Proposition 3.8. Let G be an unweighted graph. Except for (max, V V, dT (u, v)),
all max-stretch tree spanner problems coincide.
3.3.2
Anticoincidences
In order to prove for two problems that they do not coincide, we profit from the
following transitive relation: If the problems do not coincide for unweighted graphs,
then they do not coincide for graphs with metric weights. Furthermore, if there is
a graph with metric weights for which the sets of optimum solutions for two tree
spanner problems have empty intersection, then these problems cannot coincide for
general weights either. Thus, we provide the relevant anticoincidences by moving
from the most specialized weight function to general weight functions.
Unit Weights. As by Proposition 3.8 there are only two different maximum stretch
tree spanner problems in the case of unweighted graphs, we only have to establish
one single anticoincidence.
Example 3.9 (MMST vs. MDST). Consider the unweighted simple graph G in Figure 3.2(a). Recall from Proposition 3.8 and from Theorem 3.4 that in the unweighted
case we may think of the MMST problem as (max, E \ T, dT (u, v)). Hence, we are
looking for a spanning tree whose non-tree edges are linked by paths in T whose
maximum length is minimal. The spanning tree that we highlight in Figure 3.2(b)
attains an objective value of two. Moreover, every spanning tree that attains an
objective value of two has to induce all five triangles of G as its fundamental circuits.
Thus, such a spanning tree must contain the four edges that are not incident with
the infinite face. So it must not contain the edge e.
In contrast, for that in the MDST problem a diameter of three can be achieved,
the leftmost vertex and the rightmost vertex have to be connected via a path of three
edges. Observe that there is only one such path. But this includes the edge e, see
Figure 3.2(c) for one of the two optimum trees.
100
(a)
(b)
(c)
Figure 3.2: An unweighted graph and example trees which show that MMST and
MDST do not coincide
Fortunately, there exists a fairly small graph with metric weights such that the unique
optimal solutions for these three problems are disjoint.
Example
3.10 ((max, E \ T, dT (u, v) + w(e)) vs. (max, E \ T, dT (u, v))
(u,v)
). Consider the graph in Figure 3.3(a). In Table 3.1 the
vs. max, E \ T, dTw(e)
objective values of the three spanning trees in Figures 3.3(b)3.3(d) with respect to
the three objective functions are collected.
5
3
4
3
(a)
4
3
(b)
4
3
(c)
4
3
1
3
(d)
Figure 3.3: A graph with metric weights and its different optima with respect to the objective functions (max, E \ T, goal), where goal {dT (u, v) +
(u,v)
w(e), dT (u, v), dTw(e)
}
There are precisely seven circuits in G. One can easily check that the values 10
and 6 are the best values with respect to the objective functions dT (u, v) + w(e)
and dT (u, v), respectively, even when considering arbitrary sets of three circuits.
Finally, performing a simple inspection of the few relevant cases one can further
check that no other spanning tree achieves better values with respect to the three
objective functions.
101
Table 3.1: The values with respect to the different objective functions for the spanning trees in Figures 3.3(b) to 3.3(d)
Tree
Figure 3.3(b)
Figure 3.3(c)
Figure 3.3(d)
dT (u, v) + w(e)
10
11
12
dT (u, v)
7
6
7
dT (u,v)
w(e)
7
3
6
3
2
Our tour through the average stretch tree spanner problems follows the trace of our
expedition through the maximum stretch tree spanner problems. But we will find
many more different problems in the average stretch case.
3.4.1
Coincidences
Comparing the maximum stretch case to the average stretch case on general weights,
metric weights, or unit weights, the number of different problems is by up to four
larger for average stretch tree spanners.
General Weights. There are only two pairs of average stretch tree spanner problems that coincide for general weights.
P
Proposition
3.11. The two average stretch problems ( , E, dT (u, v) + w(e)) and
P
( , E, dT (u, v)) coincide.
P
, E \ T,
dT (u,v)
w(e)
Metric Weights. Much similar to the case of maximum stretch tree spanners, for
metric weight functions four problems whose objective functions involve fractions
coincide.
102
Proposition 3.13. Let (G, w) be an undirected graph with a metric weight func(u,v)
tion w on the edges. Let domain be either E \ T or E, and let term be one of dTw(e)
P
T (u,v)
and ddG
(u,v) . Then, the four problems ( , domain , term ) coincide.
Proof. In the case of a metric weight function w on the edges, for every edge e =
{u, v} E there holds dG (u, v) = P
w(e). Hence, for each of the two domains that we
consider here, the two problems ( , domain, ) coincide.
(u,v)
= 1. Thus, for
Moreover, for every tree edge e = {u, v} T there holds dTw(e)
every spanning tree its objective value with respect to the domain E is precisely n1
greater than the objective value with respect to the domain E \ T .
Unit Weights. With the exception of the average stretch tree spanner problems
that are defined for V V , all other average stretch tree spanner problems coincide
on unweighted graphs. Similarly to (3.8)(3.10) we find,
X
X
dT (u, v) + w(e) =
dT (u, v) + w(e) + 2(n 1)(3.11)
e={u,v}E
e={u,v}E\T
e={u,v}E
dT (u, v) + m
e={u,v}E\T
dT (u, v) + m + n 1.
(3.12)
(3.13)
Again, we profit from the fact that for every edge e = {u, v} there holds dT (u, v) =
dT (u,v)
dT (u,v)
w(e) = dG (u,v) . Together with (3.11)(3.13) we conclude
Proposition 3.14. Let G be an unweighted
P graph. Then the
P following eight unweighted tree spanner problems coincide: ( , E \ T, ) and ( , E, ).
3.4.2
Anticoincidences
In the case of average stretch tree spanner problems, it will turn out that even for
the unweighted case, both problems with domain V V do not coincide with any
other problem.
Unit Weights. In the case of the most special weights, the following Example 3.15
shows that we remain with 3 problems.
P
T (u,v)
Example 3.15 (MSFCB vs. STPLST vs.
, V V, ddG
(u,v) ). Consider the unweighted planar graph G in Figure 3.4. Observe that the graph from Figure 3.2
can be obtained from G simply by contracting one single edge. Again, the unique
minimum cycle basis of G consists of the five circuits which are the boundary of the
finite faces of G. Hence, the optimum solution value of the MSFCB problem on G
103
is 16 and it can be obtained by the fundamental circuits that are induced by eight
spanning trees, one of which we display in Figure 3.4(b). These eight spanning trees
all contain the four edges of G which are not incident with the infinite face of G,
230
and
P yield objectivevalues of at least 66 and 6 for the STPLST problem and for
T (u,v)
, V V, ddG
(u,v) , respectively.
(a)
(b)
(c)
(d)
value of 227
6 . But for the objective functions of MSFCB and STPLST these trees are
suboptimal because of objective values of only 17 and 63, respectively.
104
4
4
7
4
4
7
(a)
7
7
4
7
(b)
7
7
4
7
(c)
4
4
7
7
(d)
P
value with respect to ( , E, dT (u, v)) than the former trees did. Indeed, the spanning
tree T that we display in Figure 3.5(d) yields an objective function value of only 56.
One can easily observe that T is the unique minimum spanningPtree of (G, w). Actually, it is even the unique optimum solution with respect to ( , E, dT (u, v)).
P
P
P
(u,v)
Example 3.17 (
, E \ T, dTw(e)
vs. {( , E \ T, dT (u, v) + w(e)), ( , E \ T,
P
dT (u, v)) , ( , E, dT (u, v))}). Consider the graph G of Figure 3.6(a). Because of the
very regularly structured weights we need only to consider two families of spanning
trees: those that include the edge of weight 9, and those which do not. Within both
families then all trees constitute indistinguishable solutions for all the considered
problems. Representatives for the families are depicted in Figures 3.6(b) and 3.6(c),
respectively. The following
table now
P
proves the desired claim: whereas for the
dT (u,v)
, E \ T, w(e)
fractional problem,
it does not pay off to include the expensive
edge of weight 9, it does for the other three problems. It rather turns out to be
good to include this particular edge such that it can be used as a shortcut when
considering dT (u, v) for (u, v) = e E \ T .
General Weights. At last we need to consider graphs with non-metric weight
functions to prove the remaining anticoincidences.
P
P
P
dT (u,v)
(u,v)
T (u,v)
vs. {
,
, E \ T, ddG
,
E,
Example 3.18 (
, E \ T, dTw(e)
(u,v)
dG (u,v) }).
Consider the graph with non-metric weights in Figure 3.7. A first observation is that
the edge with weight 8 is not metric. More important, the edge with weight 1 is
included in every optimal tree for all the three problems. Otherwise we immediately
have a contribution of 14which is the length of a shortest circuit through this
105
7
9
7
7
9
7
(a)
7
9
7
(b)
(c)
Table 3.3: The objective values of the spanning trees in Figure 3.7 w.r.t. the three
1
problems of Example 3.18. A term of 840
is factored out for clarity reasons.
P
P
P
(u,v)
dT (u,v)
T (u,v)
Tree
, E \ T, ddG
,
E,
, E \ T, dTw(e)
(u,v)
dG (u,v)
Figure 3.7(b)
Figure 3.7(c)
Figure 3.7(d)
2555
2583
3612
(a)
(b)
5240
5208
6252
6
1
8
(c)
6
1
6
1
2720
2688
3612
8
(d)
Figure
3.7: A weighted
Figures (b) and (c) show
graph and example trees. P
P
dT (u,v)
T (u,v)
, E \ T, ddG
does neither coincide with
nor with
, E \ T, w(e)
that
(u,v)
P
T (u,v)
, E, ddG
(u,v) .
106
P
P
dT (u,v)
T (u,v)
Example 3.19 (
vs.
, E \ T, ddG
,
E,
(u,v)
dG (u,v) ). We will show the anticoincidence of the two tree-spanner problems with the help of the weighted graph (G, w)
in Figure 3.8(a). The dots in the figure shall indicate that we assume a sufficiently
large number of clips, i.e., 4-circuits that share one common edge e or f , respectively.
Let M denote this number. Further, we will refer to the edges with weight equal
to 101 as clip-edges.
Recall from Proposition 3.13 that the two problems coincide in the case of metric
weights. Hence, we chose the weight function w such that precisely one edge is
not metric: the edge g. To show the anticoincidence we will argue as follows: in
the beginning we show that each spanning tree T that is optimal for any of the
two problems must have a certain structure. First, all edges having weight one are
included in T , second, T does not contain any clip-edge, and third, the edges e and f
are in T . See Figure 3.8(b), where we highlight edges that have to be in T . Edges
that are not in T are depicted by dotted lines in this figure.
Observe that as we obtain this structure for parts of the graph where all edges
are metric, the structural properties apply to the optimum solutions subject to both
objective functions that we are investigating
having
Thereafter,
P
P in this example.
dT (u,v)
dT (u,v)
, E, dG (u,v)
this common structure, optimal trees to
, E \ T, dG (u,v) and
become distinguishable on the remaining part of the graph, where the only nonmetric edge g is going to play a key role.
So, we start motivating the mentioned structure of an optimal tree T . We first
show that w(a) = 1 implies a T . Notice first that for any of the 2M clips at least
one edge of the clip with weight one has to be in T because otherwise the tree T
would not be connected. Hence, assume that for a clip exactly one edge of weight
one and its clip-edge are contained in T . In that case, however, we get immediately
a contradiction to the optimality of T : A simple exchange of the clip-edge by the
non-tree edge with weight one within this clip instantly effects a better tree. To see
this, observe that for no other pair of vertices in G the corresponding path in T can
T (x,y)
traverse one of these two edges and compare the according values ddG
(x,y) .
Now, we know w(a) = 1 a T . Next, we establish that w(a) > 100 implies a 6
T in any tree that is optimal with respect to one of the two objectives that we are
investigating in this example. To this end, consider one bundle of clips, say the one
that contains the edge e, and assume an optimal tree T contains a clip-edge. Since
we already know that the edges having weight one are in T , the tree can contain at
most one clip-edge, because otherwise the tree T would include a cycle. Similarly,
we conclude that e 6 T . Again, such a structure contradicts the optimality of T ,
because another local change on T improves the tree: This time we exchange the
clip-edge c that we assume to be contained in T with the edge e and obtain a different
tree T = T {e} \ {c}. This exchange shortens the length of the path in T between
the vertices incident to e, thereby shortening the distances of all paths within T that
T (x,y)
contain these two vertices. In addition, even when comparing the value ddG
(x,y) of the
non-tree edge e w.r.t. T with the corresponding value of the non-tree edge c w.r.t. T
an improvement is obtained. Notice that here we only define the particular tree T to
contain the edge e. But so far nothing is said whether e is contained in any optimum
107
1 1
...
100
24
66
(a)
1 1
101
101
101
100
101
100
1
101 1
100
...
100
1
1
...
101
101
101
e
g
...
v
(b)
Figure 3.8:
A weighted
P graph on
which the optimum solutions for
P
dT (u,v)
T (u,v)
, E \ T, ddG
and
do not coincide. Whereas for the first
,
E,
(u,v)
dG (u,v)
objective it pays off to include
the non-metric edge g into an optimal tree an optimal
P
dT (u,v)
solution to
, E, dG (u,v) is attained without the edge g.
tree.
The last structural property that we are about to develop for an optimal tree T
with respect to any of the two objective functions, is {e, f } T . We already know
that T contains all edges of weight one and no clip-edge. Hence, at least one of
the edges e and f is in T , because otherwise the tree T was disconnected. So we
assume for contradiction without loss of generality that e T but f
/ T . Then,
consider the unique path P between the vertices u and v in T where obviously f 6 T
implies e 6 P . Now, an exchange of any of the edges of P by the edge f will lead to
a contradiction to the choice of T , which was an optimal tree. One can see this as
follows: before the exchange, each f -clip-edge induces a path in T of length dT at
least 126. Therefore, in both objectives each f -clip-edges contributes at least M 126
101 .
102
. It is
After the exchangei.e., now with f T this amount decreases to M 101
clear that we may choose the parameter M so large that this gain compensates the
possibly appearing increases of contributions of the remaining part of the graph that
is independent of M .
This way we force {e, f } T , and in a sense decouple the clip-edges from the
remainder of the tree.
At this point we developed all of the structural properties of an optimal tree T .
Let us emphasize that the properties hold for optima of both objectives, because up
to this point we only argued for parts of the graph on which the two objective values
differ by a constant term, because the edges within the clips are all metric.
For the remainder of the graph we discuss the effect of adding two more edges
to our tree T . Observe that there are exactly two spanning trees that contain the
edge g and three which do not.
We start by computing the objective value K that the three non-tree edges that
are distinct from clip-edges contribute. If g T , then K = 491
33 [14.87, 14.88].
108
In this section we aim at detecting anticoincidences between max-stretch and averagestretch problems. Therefore we consider unweighted versions of the problems.
Example 3.20 (MMST vs. MSFCB). Consider the unweighted simple 2-vertex
connected undirected planar graph G in Figure 3.9. We will argue that an optimum
solution for the MSFCB problem contains the edge e, whereas an optimum solution
for the MMST does not.
e
f
(a)
e
f
(b)
f
(c)
109
e
v
(a)
e
v
(b)
(c)
110
Recall that whenever a spanner problem (goal, domain, term) is NP-hard in the
unweighted case, it is in particular NP-hard in its weighted versions, too. We are
aware of three such negative complexity results for unweighted tree spanner problems.
Theorem 3.23 ([JLK78]). The STPLST problem is NP-hard.
Theorem 3.24 ([DKP82]). The MSFCB problem is NP-hard.
Theorem 3.25 ([CC95]). The MMST problem is NP-hard.
There have even been identified special classes of graphs on which these problems
are still NP-hard. For instance, think of the MMST problem on planar graphs
(see [FK01]), on chordal graphs ([BDLL04]), and on chordal bipartite graphs (see
[BDL+ 03]). But there is also one positive complexity result that has been obtained
for a tree spanner problem.
Theorem 3.26 ([HT95]). The weighted MDST problem can be solved in O(mn +
n2 log n) time.
Now, the UNTS provides us with a clear perspective on 20 tree spanner problems. In particular, Examples 3.15, 3.22, and 3.21 establish
that none of the above
P
T (u,v)
, V V, ddG
complexity results apply to the problem
(u,v) . Fortunately, we are
able to settle its complexity status in Section 3.6.1 by establishing NP-hardness.
Moreover, by Proposition 3.8 we know that in the
case of unweightedgraphs the
T (u,v)
classical maximum stretch tree spanner problem max, V V, ddG
coincides
(u,v)
with (max, E \ T, dT (u, v) + w(e)). In Section 3.6.2 we compare two inapproximability results that have been obtained for these problems. Interestingly, the result
which had never been stated before in the language of tree spanners turns out to be
stronger.
111
3.6.1
P
, V V,
dT (u,v)
dG (u,v)
is
Proof. As our proof is similar to the proof for Theorem 3.23 as it has been given
by Johnson et al. ([JLK78]), we adopt the notation of [JLK78]. We consider the
following problem, which is known to be NP-hard (Problem SP2 in [GJ79]):
Exact Cover By 3-Sets (X3C): Given a family S = (1 , . . . , s ) of
3-element subsets of a set T = (1 , . . . , 3t ). Does there exist aSsubfamily S S of sets with pairwise empty intersection, such that S =
T?
Given an instance I of X3C, we define an unweighted simple undirected graph G =
(V, E) as follows, see Figure 3.11 for an example:
R = {0 , 1 , . . . , r }, where r := |T |2 + 2 |S| |T | + 1, R0 = {0 }, R = R \ R0 ,
V = R S T,
E = {{i , 0 } : i = 1, . . . , r} {{0 , } : S} {{, } : S},
...
1 , . . . , r
0
S
2 3 4 5 6 7 8 9
P
T (u,v)
, V V, ddG
Figure 3.11: The instance of
that we associate with the in(u,v)
stance {(1, 2, 3), (3, 4, 5), (4, 5, 7), (6, 8, 9), (7, 8, 9)} of X3C
T
Denote by X the number of pairs of elements of T that are not contained together
in any of the sets of S, i.e.,
X := | {(1 , 2 ) V V : S : 1 6 or 2 6 } |.
P
T (u,v)
We will prove that
, V V, ddG
(u,v) has a solution of value at most
|V |2 |V |
|T |2 + 6 |S| 5 |T | X,
(3.14)
(3.15)
112
P
, V V,
dT (u,v)
dG (u,v)
is a star tree.
Claim. In Table 3.4, we investigate the distances between any pair of nodes in G, in
an arbitrary star tree F , and in an arbitrary non-star tree F .
In the case of a non-star tree F , there exists one vertex s S such that dF (i , s )
dF (i ,s )
F (i ,s )
4 for all i = 1, . . . , r. In particular, ddG
(i ,s ) 2, whereas dG (i ,s ) = 1, cf. the ()entry in Table 3.4. Hence, when considering R S, by the choice of r the objective
value of F is by at least |T |2 + 2 |S| |T | + 1 larger than the one of F .
Table 3.4: Distances between pairs (u, v) of nodes within the graph. The first column
denotes the cardinality of the considered subset of V V .
set of u set of v number of node pairs
R0
R
|T |2 + 2|S| |T | + 1
R0
S
|S|
R0
T
|T |
R
R
(|T |2 + 2|S| |T | + 1)
(|T |2 + 2|S| |T |)
2
R
S
(|T | + 2|S| |T | + 1) |S|
R
T
(|T |2 + 2|S| |T | + 1) |T |
S
S
|S| (|S| 1)
S
T
|S| |T |
T
T
|T | (|T | 1) X
T
T
X
dG (u, v)
1
1
2
dF (u, v)
1
1
2
dF (u, v)
1
1
2
2
2
3
2
{1, 3}
2
4
2
2
3
2
3
4
4
2
()
3
2
1
2
4
+ 2,
dG (u, v)
dG (u, v)
and for (u, v) T T there holds
dF (u, v)
dF (u, v)
+ 1.
dG (u, v)
dG (u, v)
Hence, any non-star tree F can only gain |T |2 + 2 |S| |T | on S T and T T ,
which is strictly smaller than its loss on R S.
P
T (u,v)
Now that we know that an optimum solution to
, V V, ddG
(u,v) is always a
star tree, we will compute the objective value of an arbitrary star tree F . According
to Table 3.4, it remains to investigate in detail pairs of vertices from the sets S T
113
and T T . We first examine the set S T and compute for an arbitrary star tree F
X
(u,v)ST
dF (u, v)
dG (u, v)
(u,v)ST
dF (u,v)=1
dF (u, v)
+
dG (u, v)
(u,v)ST
dG (u,v)=1, dF (u,v)=3
(u,v)ST
dG (u,v)=3
dF (u, v)
+
dG (u, v)
dF (u, v)
dG (u, v)
= |T | + |S| (|T | 3) +
(u,v)ST
dG (u,v)=1, dF (u,v)=3
dF (u, v)
dG (u, v)
As this value is independent of F , we conclude that any two star trees differ in their
objective value only for pairs (u, v) T T .
Recall the definition of X in (3.14). Among the pairs (u, v) T T , there
are precisely X for which dG (u, v) = 4and thus dF (u, v) = 4and precisely
|T |2 |T | X for which dG (u, v) = 2. As F is a star tree, we know that dF (u, v)
{2, 4} for every (u, v) T T . Recall that the quantity X only depends on the
P
T (u,v)
, V V, ddG
instance I of X3C. Hence, a spanning tree F is optimum for
(u,v) ,
if and only if it is a star tree that maximizes the number Y of pairs (u, v) T T
for which dF (u, v) = 2. How large can Y get?
We prefer to account for the quantity Y from an alternative perspective. To that
:= F (S T ). Note that |F | = |T |, because F is
end, consider the edges FST
ST
,
a star tree. Now, we define a function p(e) for the edges e = (, ) FST
2, if |F ()| = 4,
p(e) =
1, if |F ()| = 3, and
0, otherwise.
P
Thereby, Y = eF p(e). Then the following statements are equivalent:
ST
Y = 2|T |.
, p(e) = 2.
For all e FST
and
FST
(S ) := S T.
A direct computation reveals that the total objective value of the optimum solution
for a graph corresponding to a YES-instances of X3C is right as given in Equation (3.15).
114
3.6.2
Peleg and Reshef (1999, [PR99]) prove that the MMST problem
dT (u,v)
dG (u,v)
3.7
In this chapter we presented a unified notation for tree spanner (UNTS) problems.
This allowed us to detect that several tree spanner problems coincide. This is complemented by a number of example graphs showing that no further coincidences exist.
115
However, there is still some work to do in order to draw the complete map of (in) approximability results for tree spanner problems, even more when considering more
graph classes. Nevertheless, when exploring the wide area of discrete mathematics,
we hope the UNTS to provide an accurate common language in order to prevent
double work.
4
EXPERIMENTS
With this chapter we come full circle back to Chapter 1. We conduct different computational experiments that all together help to evaluate our NSC model of Chapter 1
and its computational behavior, respectively. More precisely, we run the following
three series of experiments: first, in Section 4.1, we compare the computation performances of the three MIP solver CPLEX [CPL07], MOPS [MOP07] and SCIP [Ach07]
on selected NSC instances. Second, Section 4.2 is devoted to experiments that evaluate the influence of cycle bases on the computational behavior of the according
NSC MIP. Finally, in Section 4.3 we evaluate the NSC model that we developed in
Section 1.3. Namely, we use microsimulation in order to judge the quality of the
models solution as well as to compare the solution, i.e., the set of offsets, to existing
coordinations and to solutions obtained by other means.
4.1
The experiments in this section pursue two equally important objectives. First, by
comparing the running times of the MIPs of different sized example networks we aim
to detect the crucial size of a network, i.e., the size up to which good primal solutions
can still be found in reasonable time. Both of the terms good and reasonable time
will be specified later on. The second objective is to compare MIP solvers. This
emerges as an important task when it comes to the practical application of the NSC
models, see Section 1.4. However, a MIP solver comparison is of interest by itself,
too.
Within the experiments for comparing MIP solver we focus on one particular
aspect: the quality of upper bounds during the MIP computation. That is to say, we
evaluate the quality only of primal solutions. The quality of the lower bounds during
computationor even optimality issuesare not considered. The reason why we do
so is that in applications, such as described for example in Section 1.4, good solutions
are of primary importance; lower bounds are of secondary value. For example, when
117
118
Experiments
using the NSC MIP to optimize the coordination within the iterative optimization
scheme, as proposed in Section 1.4, traffic engineers consider solutions still to be
good that are around 4 7% off the optimum. Moreover, computation times on
the time scale of seconds are considered reasonable.
Besides the state-of-the-art MIP solver CPLEX [CPL07], we included the commercial solver MOPS [MOP07] as well as the solver non-commercial SCIP [Ach07]
in our experiments. Fortunately, we were able to perform the tests with the latest
versions of all three solvers. Namely, we used CPLEX version 10.1, MOPS with
version 8.19 and SCIP 1.0. Whereas CPLEX and MOPS come with an integrated
LP solver, we used SoPlex [Wun96] in version 1.3.2 as LP solver within SCIP. We
chose SoPlex, because we wanted to consider one non-commercial MIP and LP solver
package.
The procedure of the experiments is as follows: first, we choose example instances,
a discussion of which is located below. Then, for each of the instances we run all
three solvers with different time limits. In fact, we conduct two series of calculations:
First, we ran the MIP solvers with their default settings and, second, we used special
solver parameter calibrations that benefited the finding of good solutions. In detail,
these special solver parameters were:
in CPLEX: mipemphasis = 4
in MOPS: a composition1 of parameters regulating cuts and heuristics [Suh07]
in SCIP: a composition1 of parameters regulating cuts and heuristics [AB07].
Noticeably, however, there is no one single parameter set that benefits primal computation behavior. For example, the CPLEX parameter mipemphasis = 4 is actually
meant for finding hidden feasible solutions. Still, it brought solutions with the best
values. For tuning the solver parameters of MOPS and SCIP we followed suggestions
by Suhl [Suh07] and Achterberg and Berthold [AB07], respectively. In both cases
parameters were adjusted on the basis of our NSC instances.
Furthermore, we were not able to use uniform computer architectures on which
to run the experiments. In particular, CPLEX and SCIP run on an Intel P4 machine
with 3.2GHz, 1.0GB RAM running Linux. On the other hand, MOPS runs under
Windows XP on an Intel P-M 2.13GHz with 1.0GB RAM. This inconvenience is
due to different license requirements of the three solvers and should be kept in mind
when valuating the results in Tables 4.2 to 4.3.
As test instances, we chose six specially tailored subnetworks of the Denver network, which is real-world data provided to us by the PTV [ptv]. See Section 4.3.2 for
more information on data acquisition. The instances were built such that the whole
spectrum of network sizes is covered: we consider a small network, medium size
networks and large networks. Although all six networks originated from the same
instance, the vehicle load within the networks and the networks topologies differ.
Hence, the instances can be considered to be of sufficiently distinct characteristics.
Information on the particular networks and the according mixed-integer programs
1
119
can be found in Table 4.1. For general facts about the Denver network and the other
real-world instances we refer to Section 4.3.
Table 4.1: A list of parameters, MIP and network, respectively, for the selected instances. The abbreviation Vars hereby refers to the number of variables in the MIP.
The acronyms Box, Int, Bin stand for the different types of variables: namely,
boxed variables, integer variables and binary variables. Moreover, the number of
constraints and the number of non-zero elements, NZs, are quoted. In addition, the
number of nodes, the number of edges, and the dimension of the cycle space of the
graph, , are given, as the most important network parameters.
Param
#Vars
#Box
#Int
#Bin
#Constraints
#NZs
#Nodes
#Edges
Small
Medium 1
Denver
Medium 2
Medium 3
Big
Complete
162
90
17
10
194
412
534
294
60
33
637
1,460
634
350
78
31
752
1,708
956
520
116
60
1,128
2,572
1,500
810
168
117
1,759
4,148
2,031
1,086
251
151
2,403
5,783
19
45
27
55
147
93
67
175
109
85
260
176
121
405
285
142
543
402
The Tables 4.2 to 4.3 show the results of the experiments. The numbers displayed
there are to be read as follows: for each instance, solver, and time limit we state two
quantities. First, we give the objective value, that is, the best feasible solution up to
that time limit. Since absolute values are hard to judge, we provide for comparison
the relative gap to the optimal value or, if the optimum is not known, the relative
gap to the best known lower bounds. Information on the respectively used lower
bound are stated behind the name of the respective instance. So, notice that our
gap does not coincide with the usual gap during calculations. By usual gap we mean
the following: instead of using the formula 100 (U Bt LBt ) / U Bt 2 to calculate
the gap at time t, where U Bt , and LBt denote the best upper and lower bound
at time t, we rather use a global lower bound LB that is independent of t for the
gap calculation: 100 (U Bt LB) / U Bt . For all other instances with the exception
of Denver Big and Denver Complete we took the optimal value as LB. For the
instances Denver Big and Denver Complete, we set LB to the value of the lower
bound obtained by a CPLEX run of 3 hours with an emphasis on moving best
bound. We decided to introduce this special gap since it better evaluates the primal
solutions, which we are exclusively interested in.
In the remainder of this section we will analyze the results depicted in Table 4.2,
2
Notice that this is only true for CPLEX and MOPS. The solver SCIP uses the formula 100
(U Bt LBt ) / LBt instead.
120
Experiments
CPLEX
MOPS
Denver Small
2s
179.42
0# %
LB = 179.42
179.42
Denver Medium 1
2s
5s
10s
30s
60s
600s
1072.87
1073.05
1024.23
1023.28
1023.28
.
4.6%
4.6%
0.1%
0%
0# %
.
1310.22
1294.18
1257.92
1291.80
1291.80
1246.43
4.9%
3.7%
1.0%
3.6%
3.6%
0.0%
1281.66
1281.66
1281.66
1051.74
1051.74
1025.00
1943.76
1943.76
1922.35
1922.08
1914.14
1914.14
2.3%
2.3%
1.3%
1.2%
0.8%
0.8%
2001.67
1855.01
1855.01
1855.01
1855.01
1252.53
2s
5s
10s
30s
60s
600s
9625.37
8595.51
8595.51
7816.67
7633.61
7391.64
38.2%
30.8%
30.8%
23.9%
22.1%
19.5%
12729.10
12299.62
11703.80
10761.32
10577.32
10116.24
35.4%
33.1%
29.7%
23.6%
22.2%
18.7%
20.2%
20.2%
20.2%
2.7%
2.7%
0.2%
1533.67
1533.67
1226.22
1043.71
1023.28
33.3%
33.3%
16.6%
2.0%
0%
37.8%
32.8%
32.8%
32.8%
32.8%
0.5%
2372.54
2258.01
1782.19
1528.98
1303.75
47.5%
44.8%
30.0%
18.5%
4.4%
37.7%
37.7%
37.7%
37.7%
37.7%
6.1%
8910.22
6434.31
3261.79
78.7%
70.5%
41.8%
LB = 5948.01
16176.12
16176.12
16176.12
16176.12
16176.12
Denver Complete
2s
5s
10s
30s
60s
600s
0# %
LB = 1898.22
3048.62
3048.62
3048.62
3048.62
3048.62
2021.82
Denver Big
179.42
LB = 1245.90
Denver Medium 3
2s
5s
10s
30s
60s
600s
0# %
LB = 1023.28
Denver Medium 2
2s
5s
10s
30s
60s
600s
SCIP
63.2%
63.2%
63.2%
63.2%
63.2%
LB = 8225.63
121
CPLEX
MOPS
Denver Small
2s
179.42
0# %
LB = 179.42
179.42
Denver Medium 1
2s
5s
10s
30s
60s
600s
1027.15
1024.85
1024.41
1023.28
1023.28
1023.28
0.4%
0.2%
0.1%
0%
0%
0%
1347.38
1252.60
1250.90
1250.90
1248.25
1245.90
7.5%
0.5%
0.4%
0.4%
0.2%
0%
1179.46
1054.66
1054.66
1054.66
1054.66
1028.64
2141.14
1911.79
1911.79
1900.68
1899.23
1899.23
11.4%
0.7%
0.7%
0.1%
0%
0%
2273.52
1744.24
1744.24
1438.15
1438.15
1262.06
2s
5s
10s
30s
60s
600s
8688.15
8042.08
7597.81
7500.11
7459.78
7324.00
31.5%
26.0%
21.7%
20.7%
20.3%
18.8%
11646.70
10761.68
10761.68
10390.12
10212.82
9871.35
29.4%
23.6%
23.6%
20.8%
19.5%
16.7%
13.2%
3.0%
3.0%
3.0%
3.0%
0.5%
1574.80
1394.01
1373.78
1054.73
1033.63
1023.28
35.0%
26.6%
25.5%
3.0%
1.0%
0%
45.2%
28.6%
28.6%
13.4%
13.4%
1.3%
2070.04
1773.93
1517.89
1448.98
1448.98
1265.92
39.8%
27.8%
17.9%
14.0%
14.0%
1.6%
41.2%
35.0%
24.7%
10.4%
5.6%
1.2%
3233.54
2994.85
2994.85
2477.02
2246.92
2122.15
41.3%
36.6%
36.6%
23.4%
15.5%
10.6%
10279.69
10279.69
9593.97
9593.97
9143.68
42.1%
42.1%
38.0%
38.0%
35.0%
17295.78
13306.97
52.4%
38.2%
LB = 5948.01
16037.18
13777.35
10179.80
8140.82
8140.82
Denver Complete
2s
5s
10s
30s
60s
600s
0# %
LB = 1898.22
3226.78
2918.32
2520.52
2118.95
2010.65
1921.14
Denver Big
179.42
LB = 1245.90
Denver Medium 3
2s
5s
10s
30s
60s
600s
0# %
LB = 1023.28
Denver Medium 2
2s
5s
10s
30s
60s
600s
SCIP
18458.52
15265.57
13010.12
10851.25
62.9%
56.8%
41.6%
26.9%
26.9%
LB = 8225.63
55.4%
46.1%
36.8%
24.2%
122
Experiments
results with solvers running with default settings, and in Table 4.3, results with
solvers running with tuned settings.
It is not at all surprising that in all instances and for all time limits CPLEX outperforms MOPS and SCIP. Especially with default solver settings, CPLEX outpaces
the other two solvers that for the larger instances have severe difficulties to find a
feasible solution at all.
When comparing MOPS and SCIP with their default settings, the following attracts attention. First, their performance is quite comparable in the first, i.e., smallest, three instances. For the larger instances, however, SCIP fails to provide feasible
solutions. Here, MOPS performs a little better, that is to say, comes up with feasible
solutions. Nevertheless, they are mostly of poor quality.
In this context, the following is remarkable: we defined the NSC problem, see
Section 1.2.1, using node offsets. So, when modeling the NSC with node offsets,
finding a feasible solution should be unproblematic, since any solution is feasible.
However, because we decided to use link offsets for our NSC MIP (1.25), this is
no longer the case. Nevertheless, although not explicitly stated in our model, the
Equations (1.2) that couple the node offsets and the link offsets are inherent in
the NSC MIP (1.25). They propose a trivial way to find a feasible solution: for
example, the all zero node offset vector can be easily transformed into the link offset
vector. Hence, the fact that MOPS and SCIP for the larger instances fail to come
up fast with a feasible solution, shows that these solvers cannot detect the implicit
connection (1.2).
Now, let us remember that for the two largest instances, Denver Big and Denver
Complete, we do not know the value of an optimal solution. Therefore, we had to
use a rather poor lower bound. Nevertheless, we think that the gaps in fact are
misleading, and the obtained solutions are closer to the optimum value than the
percentages reveal.
When considering Table 4.3 with the results for the runs with tuned parameter
settings one notices the following. All three solvers perform better compared to
the runs with their default settings. The difference becomes apparent especially for
MOPS and SCIP. With the tuned parameter settings feasible solutions are found for
all instances, albeit not as fast as CPLEX.
In the beginning of this section, we raised the question of the crucial size of a
network. When analyzing the results of the experiments we observe that providing
good solutions in reasonable time for medium size instances is not problematic for
the state of the art MIP solver CPLEX. In fact, we believe that even for Denver
Big and Denver Complete the values in Tab. 4.3 of solutions obtained by CPLEX
are of satisfying quality, too. Thus, we conclude that finding good solutions for
real world NSC instances of sizes up to approx. 150 signals is possible. As for
the other solvers, MOPS and SCIP, the results show that small to medium size
networks are tractable. Nevertheless, with a more evolved solver parameter tuning
both MOPS and SCIP may perform better. During the experiments both solvers, but
especially SCIP, showed a remarkable sensitivity towards the adjustment of the solver
parameter settings. Notice, though, that discussion of a crucial size of a network
only refers to the application of the model within the optimization scheme proposed
123
In this section we conduct experiments that explore the influence of cycle bases on
the performance of a mixed-integer linear program. That is to say, we build for
selected example networks the according NSC formulation using different types of
cycle bases and evaluate the MIP performance. First we discuss the reasoning behind
this kind of experiment and the experimental settings, i.e., the chosen criteria for
measuring the quality of a cycle basis and measuring the MIP performance, the
selected instances and cycle bases.
For many real-world applications cycle bases are needed to build appropriate
models. That means, often when modeling problems via a graph, a structural property has to hold for all cycles in a graph. Obviously, one does not want to formulate
constraints for all cycles. Then it is equivalent, to formulate the needed constraints
for a subset of the cycles: a cycle basis. It can be shown that, for several applications, different cycle bases are in use to maintain a concise problem formulation, see
Chapter 1 and [Gle01, Lie06, Bol02].
However, since there is a huge number of different cycle bases, one question naturally arises: Are all cycle bases equally good, or do some bases result in a better
problem formulation? The possible reasoning behind this question is the following: when we look at the problem formulation, often cycle bases are used to model
some kind of periodicity. Then, as it can also be observed in our NSC MIP (1.25),
those periodicity constraints demand integer variables for which one wants to give
124
Experiments
good bounds; naturally, good bounds immediately reduce the search space and induce a correspondingly tighter problem formulation. For example, Equations (1.9)
and (1.10) show how bounds for the NSC model can be obtained. Here, of course,
different cycle bases yield different bounds on the integer variables. The hope is
that cycle bases that allow for tight bounds, and thus result in a tighter problem
formulation, lead to MIPs that perform better.
Although the above reasoning is neither surprising nor new, there are hardly
no studies about the real influence of bases on the performance of the according
MIP. Only Liebchen [Lie06] reports on that topic in the context of modeling periodic
timetables.
We chose as test instances three subnetworks of the Denver network with different
sizes, namely the instances Denver Medium 3, Denver Big and Denver Complete. For
data on the networks and the according mixed-integer linear programs see Table 4.1.
Notice, that the data given there does not depend on the selected cycle basis.
A canonical quality measure of a cycle basis when trying to anticipate the computational behavior of its MIP is the width W (B) of a cycle basis B, see [Lie06]. It
is defined as
Y
n n + 1 .
W (B) :=
(4.1)
B
Hereby denote n and n an upper and lower bound, respectively, on the integer
variable n of the NSC MIP constraint of the cycle B. See Equations (1.9)
and (1.10) in Section 1.3.3.
However, just as Liebchen suggests in [Lie06], we use
W (B) := log W (B) =
log(n n + 1)
(4.2)
since this way we avoid the uncomfortable product in (4.1). Notice that replacing W (.) by W (.) is valid, because the log-function is monotonically increasing. The
importance of the width of a cycle basis is nicely summarized by Liebchen who concludes that the width of a cycle basis suggests how CPLEX might perform on the
resulting integer program [Lie06]. One aim of the experiments in this section is
to investigate to what extent this observation can be shared. It is worth mentioning, however, that other quality measures of a cycle basisfor example its length,
w.r.t. a certain cost function or w.r.t. the total number of edgesare possible, too.
We decided for the width of a basis as measure because it has the most immediate
connection to the MIP, as it predicts its search space.
Another crucial question is how the influence of a cycle basis on the performance
of the according MIP can actually be determined. That is to say, to which MIP
measure we do relate the width W (B) or W (B), respectively, of the considered
basis B? Among possibilities like total computation time or value of the according
LP relaxation we decided to take the lower bound after 10 seconds as criteria. We
did this for the following reason. The value of the LP relaxation has the major
shortcoming of having a very narrow range of values. Namely, we observed a range
of approximately 2.5% within all occurring LP relaxation values for experiments on
125
the three considered instances. On the other hand, considering the total computation
time was not possible for the selected instances. In addition, we feel that considering
the lower bound of the MIP (after a short time of computation) better reflects the
pure connection between the cycle basis and its MIP since it blanks out primal
heuristics. The choice of the time limit of 10 seconds then was also a concession to
the great number of single experiments that we were about to conduct. Other values
for the time limit are possible, however.
In the next paragraph we specify which cycle bases we considered for our experiments. We chose two specific single bases and two series of bases. Namely we
examine:
1. a strictly fundamental cycle basis that is induced by a Minimum Spanning Tree
of the graph with the negative of the edge multiplicities as cost function on the
edges. In Fig. 4.2 this tree is denoted by B1.
2. A second single basis denoted by B2 that is built from B1 by the following
heuristic: Consider the spanning tree T that induces B1. Further, sequently
consider branches e
/ T . If e induces a cycle of length 2 or has no parallel edges
do nothing. Otherwise, substitute for each edge f
/ T parallel to e its induced
cycle of length 3 by the length 2 cycle consisting of the edges e and f . See
Fig. 4.1 for an example. It is obvious that then B2 is a basis. Moreover it can
be shown that B2 is weakly fundamental and hereby integral.
3. Furthermore we consider a series of 500 bases induced by random spanning
trees. We constructed a random spanning tree by putting random integer
costs (between 1 and 103 ) on the edges and computing a minimum spanning
tree. However, for parallel edges we chose the same random cost. We do so,
because otherwise edges with high multiplicity are likely to be in the tree and,
hence, the induced basis is too similar to the one described in 1. In Fig. 4.2
these trees are collected under rand-ST.
4. Finally, we take the very same 500 bases as in 3 as starting bases and refine
them individually with the heuristic described in 2. In Fig. 4.2 these trees are
denoted by WFCB-heuristic.
Notice that the bases described in 1 and 3 are strictly fundamental whereas those
from 2 and 4 are only weakly fundamental. All bases considered are integral.
We conclude the description of the experimental settings by mentioning that
the experiments were performed on an Intel P4 machine with 3.2GHz, 1.0GB RAM
running Linux. As MIP solver we used CPLEX in version 10.1 [CPL07].
In Figures 4.2(a) to 4.2(c) the results of the experiments are collected and they
will be discussed in the remainder of the section.
As can be seen from the diagrams 4.2(a) to 4.2(c) the obtained results are consistent with respect to the different instances. In all three cases we observe that the
random spanning trees induce bases with a much greater width than their weakly
fundamental counterparts; this, however, is not surprising. Neither it is surprising
that basis B1 has a far smaller width than the average strictly fundamental one.
126
Experiments
e7
e7
e8
e5
e6
e2
e1
e1 e5
e4
e3
(a) G
(b) T
e7
e8
e7
e5 e6
e2 e1
e5 e1
e4
e7
e8
e5 e6
e2 e1
e4
e3
After all, B1 was built such that edges with high multiplicity are likely to be in
the tree, which, in turn, results in many short, i.e., length 2, cycles. In addition,
the following fact stresses this observation: when considering a minimum spanning
tree with positive multiplicities as costs on the edges one gets the following three
tuples of (W (B), lower bound after 10 seconds of MIP computing ) for the three instances: (291.26, 991.34), (471.56, 4692.88) and (657.65, 6742.40). The numbers show
that these bases have a huge width and, not forestalling too much, inferior lower
bounds compared to the considered bases. For clarity reasons, we omitted theses
bad examples in Fig. 4.2.
In considering the results plotted in Figs. 4.2(a) to 4.2(c) our conclusion is twofold.
First, and unfortunately, a clear straight correlation of the width of a basis to the
MIP performance, i.e., the MIP lower bound after 10 seconds of computation, can
not be affirmed. Namely, a basis with a smaller width does not necessarily lead to
a MIP with better, i.e., higher lower bound. Notice that this would have meant
that the pairs of variates string along a monotonically decreasing function. However,
this can not be observed in the diagrams. For example, for all three instances there
are pairs of bases whose widths are almost equal but whose resulting lower bounds
diverge significantly. See for example in 4.2(a) bases with tuples (251.59, 1460.99)
and (254.09, 1052.01).
Still, we draw following conclusions: when comparing the series of weakly fundamental bases to the series of random strictly fundamental ones, the weakly fundamental bases perform better. Namely, taking the average on the lower bounds
of the weakly fundamental bases, it is by 8.0%, 5.9%, and 7.4% (for the three instances) higher than the corresponding quantity of the strictly fundamental bases.
The same holds when investigating the median. The median of the lower bounds of
the weakly fundamental bases is by 8.4%, 5.5%, and 7.3% higher than the median
of the lower bounds of the strictly fundamental bases. Thus, one could interpret
this as a strong indication that improving the strictly fundamental bases, even
though they become weakly fundamental, and thereby reaching a whole smaller level
127
1550
rand-ST
WFCB-heuristic
B1
B2
1500
1450
1400
1350
1300
1250
1200
1150
1100
1050
180
190
200
210
220
230
240
log of width of cycle basis
250
260
270
280
5400
5200
5000
4800
4600
4400
4200
280
300
320
340
360
380
400
log of width of cycle basis
420
440
460
7800
7600
7400
7200
7000
6800
6600
6400
400
450
500
550
log of width of cycle basis
600
650
Figure 4.2: Diagrams that show the relation between the logarithm of the width of
a cycle basis, W (B), and the lower bound after 10 seconds of computation of the
according MIP for the selected instances and cycle bases, respectively.
128
Experiments
of widths of cycle bases, does enhance the MIP performance. Moreover, this strong
indication is stressed by the following observation. For the Denver Medium 3 network, in 94.4% of the 500 cases, the WFCB heuristic improved the obtained lower
boundcomparing the two MIPs, the strictly fundamental one and its refined version. For the other two considered instances, Denver Big and Denver Complete,
the percentages even amount to 99.4% and 99.8%, respectively. Hence, applying the
WFCB heuristic is recommended.
Finally, we discuss the initial question: to what extent can we share the observation by Liebchen [Lie06], that the width of a basis suggests how the MIP solver might
perform on the according MIP? Liebchen, who investigated a periodic timetabling
application using a PESP formulation, correlated the width of a cycle basis to other
measures for the computational performance of its MIP. Namely, he considered the
LP relaxation value and the total solution time of the MIP. He then observed that
for three independent pairs of cycle bases, always the one with a smaller width, leads
to a MIP that performs better w.r.t. the LP relaxation value and the total solution time. However, for the previously mentioned reasons, we decided to consider
the lower bound of the MIP after 10 seconds of computation as the performance
measure for our experiments. Here, an exact correlation between the width of a
cycle basis, and the value of its MIPs lower bound could not be observed. Still, by
conducting experiments comparing random SFCBs with WFCBs, which, on average,
have a far smaller width, the statement of Liebchen can be shared to a certain extend. The reason, why our experiment indicates that correlations are not as explicit
as in Liebchens experiments may be due to the underlying model. Liebchen used
an MIP approach for the PESP, whereas we modeled the NSC problem. See Section 1.2.2 for problem definitions. The fundamental difference between the PESP
and the NSC problem is, that for the PESP, one is given bounds on the periodic
tension variables. These bounds then explicitly enter the definition of the bounds of
the integer variables for the cycle equations. The bounds, in turn, are essential for
the definition of the width of a cycle basis.
We conclude this section with the following remark.
Remark 4.2. For our implementation of the NSC MIP (1.25), as it is used in Sections 4.1 and 4.3, we used the basis of type B2.
4.3
Case Studies
In this section we evaluate the NCP model (1.25) on real-world networks of Portland,
Section 4.3.3, and Denver, Section 4.3.4. Before we do so, however, we will discuss
in Section 4.3.1 the procedure of evaluating such a model and, in Section 4.3.2, the
topic of data acquisition.
4.3.1
Although models for practical problems in general or for the NSC, respectively, can
be of theoretical value, in most cases the real quality can not be determined until the
129
adten Simulation.
VISSIM abbreviates Verkehr in St
130
Experiments
in VISSIM. These stochastic functions are fed with random seeds. It is recommended
running the simulation multiple times with different seeds in order to get significant
results.
For each vehicle, the total duration of its journey from its source to its sink is
measured. Here, delay refers to the actual determined journey time minus a best
possible journey time. The best possible journey time for a vehicle is the journey
time that the vehicle needs for the same route but without any other vehicles on the
track and without signalization. That is to say, the measurement delay collects
delay due to stops because of signalization and delay due to interaction with other
vehicles. However, we think that this VISSIM definition of delay fits the one of the
MIP, see Section 1.3.
Although microsimulating a particular solution of the NSC in order to evaluate
its real-world quality is important, its real value can not be determined until other
solutions have been simulated as well. One such solution, which is important to
use for a comparison, is the present solution, i.e., the present adjusted offsets in the
network. Moreover, simulating, for example, solutions obtained by TRANSYT, see
Section 1.1.3, and comparing the simulation results to the ones obtained by our NSC
model (1.25) is of great significance.
The detailed VISSIM settings that we used for the simulations are collected in
the following remark.
Remark 4.3. We run the simulation for 3600 seconds (Portland) and 1800 seconds (Denver) with a prior 300 seconds (both) phase to reach a stable situation in
the network. We speak of a stable situation when the number of vehicles in the
network is nearly constant. For these 300 seconds we do not measure delay. For each
solution to simulate, we run the simulation twenty times, each with different random
seed, and take the average. We keep the internal VISSIM parameter of calculation
frequency, measured in time steps per simulation second, at its default value of 10.
4.3.2
Data Acquisition
Getting real-world data for testing mathematical models for practical problems is
always a problem. This is especially true in the case of instances for the NSC; the
required data is substantial: one needs information on the network, on the traffic
flow in the network, and last, but not least, on the signalization. Fortunately, the
PTV company provided detailed data, i.e., data for the inner-city street networks
of Portland and Denver, including information on the traffic flow and signalization.
However, this reveals only half of the story.
First, the term data has to be specified, since one needs data for the definition
of an instance as input for the model and data as it can be used as input for the
simulation. Unfortunately, the two situations require different types of data. With
input for the simulation one has to know the position of each signal in the network,
i.e., its coordinates. On the other hand, this is a useless information for the NSC
model, because there we are only interested in, say, adjacency information of the
network.
131
So, the question is how to get all the necessary information; or, how can we
complete the data? In general, data for real-world traffic networks is acquired by
traffic companies, usually by order of public authorities that want some kind of model
for their network in order to plan or evaluate a certain traffic scenario. Fortunately,
the company PTV provided us with data for some north-american cities. As the
PTV is a company that develops traffic-planning software, real-world data for traffic
networks is usually preprocessed for and held within their software tools. Namely,
the software tools are VISUM4 and VISSIM. For a brief description of VISSIM see
Section 4.3.1. VISUM is a
Transportation information and planning system for private and public
transport, graphical network editing, analysis, evaluation, assignment,
forecast and impact calculations [ptv].
In Figure 4.4 the network of Denver, which is discussed in detail in Section 4.3.4,
is depicted.
(a) VISSIM
(b) VISUM
Figure 4.4: The street network of Denver as it is modeled in VISSIM and VISUM.
Actually, the two views are not independent. Rather, the VISSIM model evolved
from the VISUM one by an exportation.
As we want to use the microsimulation VISSIM for our evaluation we have to have
a VISSIM-model for the network. However, if a VISSIM-model of a network is the
only data source available, we face some other problems: within VISSIM the traffic
flow is given via inflow definitions and route decision points. As such, an extraction
of the relevant information, which is the traffic flow on a link, is complicated. In
addition, travel times on the links have to be determined via simulation. Thus,
extracting all data just from a VISSIM-model is afflicted with some effort.
However, if a VISUM-model of a network is the only data source available, the
situation is not optimal either. On the one hand, internal data structures enable a
much easier extraction of link flows and travel times. Moreover, a model-export to
4
adten Umlegung.
VISUM abbreviates Verkehr in St
132
Experiments
VISSIM is supported. On the other hand, the exported VISSIM-model often has to
be mended by hand.
So, at the best, we have a VISUM-model together with a compatible VISSIMmodel, i.e., a VISSIM-model that has copiously been repaired and tested.
Fortunately, for the network of Denver, see Section 4.3.4, the PTV provided us
with a complete VISUM-model and an associated reworked VISSIM-model. For the
network of Portland, see Section 4.3.3, we extracted all necessary information out of
an VISSIM-model for the network.
Finally, it should be noted that no data for platoon-lengths or saturation rates
was given. Instead, we used approved estimations for the platoon lengths, which
are specified in the particular sections. For the saturation rate we used a value of 1
vehicle per second.
4.3.3
Portland
The city of Portland is, though not its capitol, with a population of approx. 560,000
(2006) by far the largest city in the U.S. state of Oregon.
Figure 4.5: The inner-city street network of Portland, Oregon. Screenshot from
VISSIM.
In our study, we consider the downtown part of the inner-city street network of
Portland, cf. Fig. 4.5(a). In Fig. 4.5(b) a VISSIM screenshot of this part of the city
is displayed. Here, the blueishly highlighted streets indicate bus corridors. Along
133
these corridors traffic responsive signals are installed, whereas everywhere else we
find fixed-time control signals. This is illustrated in Fig. 4.5(c). In total, there
are 29 traffic responsive signals and 81 fixed-time control ones. For the remaining
intersections in the network, either no data about signalization is available or they are
simply unsignalized, e.g., the ones along Park Avenue (see the green stripe stretching
south west to north east in Fig. 4.5(a)).
Although it may not seem so, the VISSIM model of the street network, which we
use as database, has the disadvantage in that it has disruptions in the street layout,
whereas in the real network are none. See the black lines indicating this disruptions
in Figure 4.5(c).
Along the aforementioned bus corridor there are traffic responsive signals installed
that favor the public transport. The network is subdivided by this corridor and the
unsignalized stripe along Park Avenue into separated blocks with fixed-time signals,
cf. Figure 4.5(c).
We chose the section depicted in Fig. 4.5(d) as the part on which to run our optimization. The square between Taylor St. and Madison St. and between Broadway
and 4th Aven. consists of 16 signalized junctions, i.e., 16 fixed-time controlled signals
at the intersection. See Figure 4.5(d). Moreover, all signals work with a 54 second
cycle length and have a prescribed and fixed signal timing plan. The green phases
of the signals have lengths between 17 and 31 seconds. For the transit times on the
links a value of 10 seconds is set uniformly, since all links have the same length.
Figure 4.6 provides an enlarged view of the chosen network. In this figure we
highlight the paths on which the traffic crosses the network. The numbers indicate
the corresponding traffic volumes in number of vehicles per hour. Note that the information on paths and volumes has been experimentally extracted from the complete
Portland VISSIM model by a large number of simulation runs.
700
220
190
250
180
700
240
100
250
150
340 100
700
Figure 4.6: A visualization of the traffic leading paths crossing the network is given.
The numbers indicate the amount of traffic measured in vehicles per hour.
Due to the given data and conditions, for the optimization, we use the NSC
model (1.25) without the constraints for the variable phase sequencing, (1.25c)
and (1.25d). Thus, a solution of (1.25) for the Portland instance is a set of op-
134
Experiments
timal offsets which, because of the networks small size, could be found in less than a
second. Therefore, choosing a small cycle basis for the MIP (1.25) was of minor relevance for the Portland network. Table 4.4 summarizes some statistical information
on the considered network and its NSC mixed-integer linear program.
Table 4.4: Some collected characteristics of the Portland network and the according
NSC mixed-integer program
Param
Portland
Network:
Number of Nodes
Number of Arcs
Number of Cycles
16
30
15
MIP:
Number of Variables
Number of Constraints
Number of Non-zeros
105
133
271
Delay
in s/veh
CPU time
Relative
difference
33.7
27.0
16.6
16.1
22.2
15.9
< 1s
10s
800s
103%
63%
0%
3%
34%
4%
135
as follows. For the optimization we used CPLEX 10.1 on a Intel P4 machine with
3.2GHz, 1.0GB RAM running Linux. On the other hand, TRANSYT works under
Windows XP on an Intel T 43p 2.13GHz with 1.0GB RAM.
In the remainder of this section we discuss the results of the simulation given in
Table 4.5. A first observation is that taking just any offset, as the average result of
the random offset sets shows, is not a good idea. One risks an average delay that
is more than twice as big compared to the present coordination. Taking the best of
the 75 offsets sets does not yield acceptable results, too. These results already show
that there is optimization potential inherent in coordinating a network.
One of the main observations from Table 4.5, is that the offsets obtained by
our optimization turn out to be better than the present ones. Namely, an optimal
solution of our NSC model (1.25) results in a 3% decreased delay value. Although
this does not seem like much, one has to note the very short time that was needed
to obtain an optimal solution of (1.25): less than a second. On the other hand,
obtaining the offsets for the present coordination probably takes much longer.
Another important observation from the simulation results is that the quality of
offsets obtained by TRANSYT is twofold. Either the offsets are very good, and it
took a long time to find them, or they were found in reasonable time, but they are
of poor quality. Having both, good offsets found in short time seems impracticable.
A short explanation of the two different TRANSYT solutions is required: during
the genetic algorithm, TRANSYT evaluates the respective solutions via an internal
simulation. For that simulation two modes exist. In the first mode, the singlecycle mode, one single cycle, i.e., once a time period equal to the networks uniform
cycle length, is simulated. In the second mode, the so-called multi-cycle, a multiple
of cycle lengths is simulated. Obviously, the longer the internal simulation for the
evaluation, the more realistic. Hence the results from Table 4.5 are not surprising.
The multi-cycle approved offsets are of much better quality than the single-cycle
approved ones, 15.9 s/veh compared to 22.2 s/veh. However, this quality gain
is achieved at the cost of running time, 800s compared to 10s. All other genetic
algorithm parameters have been set equal for both modes, see Remark 1.2.
Moreover, the following has to be discussed. Although with our optimization we
found an optimal solution, another set of offsets, namely the one by TRANSYT,
performed better in simulation. The reason for that is TRANSYT uses a realistic
internal evaluation of a solution. That means TRANSYT uses a mesoscopic traffic
model, whereas in our MIP we consider the traffic macroscopically.
However, when comparing different network coordinations to the present one, we
have to admit the following. The present coordination was adjusted for the whole
Portland network and not for the 16-signal subnetwork that we selected for our
experiments. Hence, such a comparison may appear somewhat unfair. On the other
hand, an at-once coordination of the complete Portland network is tricky anyway.
This is because different parts of the network, i.e., the blocks separated by the
unsignalized corridor or the corridor equipped with traffic responsive signals, do not
interact, or, if they do, interact in an unpredictable way.
In conclusion observe from Figure 4.6 that the traffic volumes are not too high.
Namely, it is not clear whether these values completely justify the assumption of
136
Experiments
Denver
As a second real world example network we investigated the inner city street network
of Denver. The city of Denver has a population of approx. 560,000 (2005) and is the
capitol of the U.S. state of Colorado. In Fig. 4.7(a) an aerial view of downtown Denver
can be seen, whereas in Fig. 4.7(b) a VISSIM model of the network is depicted. In
total, the network consists of 146 fixed-time controlled junctions with a uniform cycle
length of 75 seconds. Unlike in the Portland network, this time we consider the whole
network, i.e., we do not tailor a smaller subnetwork.
Figure 4.7: The city of Denver, left from aerial view and right in simulation.
However, we have to discuss some preparation steps that we had to carry out.
The data for the Denver network originated from a VISUM input model. In order to
run the simulation, the network with the traffic load had to be exported to VISSIM.
Unfortunately, this sometimes caused some incorrect results. For example, connections of streets were not always copied correctly which resulted in slightly different
junction topologies. In addition, the original Denver network contained four traffic
responsive signals. At these signals, we approximated the traffic responsive controlling by a fixed-time controlling [Nok06], thereby staying very close to the original
signal settings.
Nonetheless, an advantage of having at hand a VISUM model is that traffic
volumes, as they are needed as input for the optimization, can easily be extracted.
137
Thus, unlike for the Portland network we do not have to detect traffic volumes by
extensive extra simulation runs. For clarity reasons, we do not give detailed numbers
of the traffic volumes on the links. However, for some links peak values of more
than 2000 vehicles per hour are reached. In this respect, the data is closer to what
we formerly quoted as nearly saturated conditions than the Portland network was.
Another slight modification is that we only consider links with a traffic volume greater
than a hundred vehicles per hour. This way, we reduce the network enormously and,
hence, simplify the optimization MIP. On the other hand, the model does not become
less accurate, since the edges with that low traffic volumeespecially compared to
the peak volume linksare most probably of no consequence, anyway.
Notice that this modification is the reason why the MIP and network characteristics of the Denver network given in Table 4.6 differ from those of Table 4.1.
Due to the absence of detailed data on different signal timing plans for the signals
and due to the uniform cycle length within the network we use the NSC model (1.25)
without constraints (1.25c) and (1.25d). For information on which cycle basis we used
for the MIP, see Remark 4.2. Note that this time the information is relevant, since
the model is too big to provide an optimal solution. The vehicles transit times are
obtained, as with the exact traffic volumes, by an internal VISUM calculation. As
platoon lengths we again used the values graded with respect to the traffic volumes
as proposed in 1.3.5.
Unlike for the Portland network, this time we are not able to use an optimal
offset set for the simulation. The solution that we do use for the simulation is the
best solution after 5 minutes of computing with CPLEX 10.1, with default settings,
on a Intel P4 machine with 3.2GHz, 1.0GB RAM running Linux.
Table 4.6 summarizes some statistical information on the considered network and
its NSC mixed-integer linear program.
Table 4.6: Some collected characteristics of the Denver network and the according
NSC mixed-integer program
Param
Denver
Network:
Number of Nodes
Number of Arcs
Number of Cycles
MIP:
Number of Variables
Number of Constraints
Number of Non-zeros
146
399
257
1454
1850
3801
138
Experiments
We were not able to consider offsets calculated by TRANSYT, which is a nonnegligible limitation. The reason is that we do not have a Denver model for TRANSYT at hand nor have we the possibility to export or transform data from VISUM or
VISSIM. Because of its size, we could rebuild the Portland network with its 16signals
in TRANSYT. However, this was not an option for the Denver network. First, the
network is too big to reconstruct it by hand, and second, the file format of a TRANSYT input file is rather complicated, such that we refrain from providing a script
that, for example, transforms a VISSIM input into a TRANSYT input.
Table 4.7: Comparison of delays from different coordinations.
Coordination
Average random
Best of 20 random
Present
Optimization (300s)
Delay
in s/veh
Relative
difference
192.08
171.75
128.70
132.85
49.2%
33.4%
0
3.2%
In the remainder of this section we discuss the results of the simulation given
in Table 4.7. Again, the first observation is that taking an arbitrary offset set is
not recommendable. Neither does the best of the 50 offset show acceptable results in
simulation. On the other hand, one has to observe that both values are much closer to
the present coordination as was the case for the Portland network. Therefore see, that
the gaps to the simulation result of the present coordination approximately halved.
Namely, for the Portland network, the best random offset set achieved a by 63% worse
simulation result than the present coordination, whereas this value drops to 33.4%
for the Denver network. A reason for this effect could be the following. Within the
Denver network, the paths that carry a big amount of traffic are long and rather
well distributed over the network. That is to say, there are many signals, which
are quite equally spread over the whole network and control huge traffic volumes.
Hence, the impact of random offsets may balance out over the network. However,
the optimization potential is apparent also for the Denver network.
The main observation from Table 4.7 is, though, that the present coordination
and the coordination that we calculated with (1.25) are nearly of the same level. In
particular, the present coordination achieves a delay of 128.70 s/veh in simulation
compared to 132.85 s/veh for our coordination. It is possible that this small gap is
due to the two or three preparation steps which we mention in the beginning, where
we, e.g., had to remodel some small parts of the network topology.
Unfortunately, we cannot contrast these two results with the results of offsets
by TRANSYT. Nevertheless, when considering the figures of Table 4.5, it is most
probable that finding offsets, whose quality is comparable to the present coordination,
would have lasted more than 5 minutes.
In conclusion, we think that the simulation results of our NSC models offsets,
compared to the results of the present coordination, are satisfying. Of course, pro-
139
viding offsets that outperform the present coordination would have been even better.
Nevertheless, one has to bear in mind the relatively short amount of time, 5 minutes,
that we spent finding the offsets. Although we can not say for sure, calculating the
offsets for the present coordination probably has taken much more time.
4.4
In this chapter we conducted three different types of experiments. First, in Section 4.1 we evaluated the performance of three different MIP solver, CPLEX [CPL07],
MOPS [MOP07], SCIP [Ach07], at selected real world NSC instances. We were interested in comparing upper bounds found by the three solver after a given time limit.
Both with default solver parameter settings and with individually tailored ones, that
benefit upper bound issues, CPLEX performed best. MOPS and SCIP performed
almost equally, although MOPS tended to find a feasible solution quicker. Finally,
it could be shown with these experiments, that the NSC model (1.25) can be used
for large instances, i.e., good solutions could be found in reasonable time.
In Section 4.2 we evaluated the influence of cycle bases on the performance of the
according NSC mixed-integer linear program. In order to do so, we correlated the
width of a cycle basis to the lower bound value after a computation of 10 seconds,
of the NSC MIP, that uses this particular bases. For these experiments, we basically
considered two classes of cycle bases: strictly fundamental bases and weakly fundamental ones, which we obtained by applying a heuristic to a strictly fundamental
basis. Unfortunately, a clear correlation between the two mentioned quantities
the smaller the basis width, the higher the lower boundcould not be confirmed.
Nevertheless, the study shows that applying the weakly fundamental heuristic does
decrease the basis width in such a manner that a non-negligible increase in the lower
boundon average over random start treescould be observed.
Finally, in Section 4.3 we evaluated our NSC model on the real-world instances
of Portland and Denver with the help of the microsimulation VISSIM. In both cases,
the simulation results of our model are comparable to the simulation results of the
present coordination. Namely, for the Portland network the MIP (1.25) showed
slightly better results than the present coordination, whereas for the Denver network
it was the other way round. In both cases the difference amounted to appr. 3%. Thus,
because (1.25) finds good offsets relatively fast, we conclude that the NSC model is
well-suited to optimize the coordination of fixed-time signalized networks of large
size.
Nevertheless, there remain some open questions or tasks.5 When conducting
experiments, one can think of a lot of aspects to consider. That is to say, that,
although we believe the experiments dealing with MIP solver and cycle bases of
Sections 4.1 and 4.2 were significant, one could think of the following indisputable
enhancements.
With respect to the solver studies, for example, more solvers and more instances
5
Notice, though, that open questions regarding the NSC model itself, like for example enhancements of the model are formulated in Section 1.3.7.
140
Experiments
could be included. In addition, one might think of other criteria for the comparison of
the solver. Surely, it would be interesting to include some tests on lower bounds in the
experiments, too. Furthermore, investigating whether a different model than (1.25),
e.g., a MIP that uses node offset variables, reveals different performances of the MIP
solvers, has to remain an open question.
Second, as for the study on the influence of cycle bases, one might consider more
classes of cycle bases. In addition, it would be interesting to see, whether taking
other criteria as the proposed lower bound after 10 seconds of computation confirm
our observations.
Finally, we mention a few open tasks regarding the case studies of Section 4.3.
First, a comparison of the performance of our NSC model to the quality of solutions
obtained by TRANSYT on the Denver network could not be accomplished up to
now. Surely, this comparison will provide an additional benefit to Section 4.3.4.
Generally, simulating offsets obtained by other meansfor example other traffic
software tools, or heuristicsis desirable for having a broader view, always positive
for the evaluation of a model.
Furthermore, it was unfortunately not possible to evaluate our NSC model (1.25)
in its full functionality. Neither for the Portland network nor for the Denver network
was data available that allowed a concurrent optimization of the phase sequencing.
Moreover, in both networks, the signals had a uniform cycle length. Hence, we used
the MIP (1.25) without Constraints (1.25c) and (1.25d). However, this again affects
the question of data acquisition, and we fear that real-world data that is suited to
evaluate all aspects of our NSC model is not realistic to hope for in near future.
141
Parameter settings for MOPS and SCIP, that benefit good primal
solutions
Finally, we explicitly state the special MIP solver parameters that were used in
Section 4.1 to benefit the finding of good primal solutions. For SCIP, we used the
following non-default parameter values [AB07]:
constraints/linear/maxprerounds = 0
heuristics/coefdiving/maxlpiterquot = 0.1
heuristics/crossover/nodesquot = 0.2
heuristics/fracdiving/maxlpiterquot = 0.1
heuristics/objpscostdiving/freq = 1
heuristics/rootsoldiving/freq = 10
heuristics/veclendiving/freq = 1
separating/maxroundsroot = 5
separating/flowcover/freq = 1
heuristics/coefdiving/freq = 5
heuristics/crossover/freq = 15
heuristics/fracdiving/freq = 5
heuristics/linesearchdiving/freq = 1
heuristics/pscostdiving/freq = 1
heuristics/rootsoldiving/maxlpiterquot = 0.05
separating/maxrounds = 1
separating/cmir/freq = 1
xmaxno = 3000
xlocs = 1
xmirct = 3
xheutp = 0
xnored = 2
xrimpr = 0.01
BIBLIOGRAPHY
[AB07]
[ABN07]
Ittai Abraham, Yair Bartal, and Ofer Neiman. Embedding metrics into ultrametrics and graphs into spanning trees with constant average distortion. In
Nikhil Bansal, Kirk Pruhs, and Clifford Stein, editors, SODA, pages 502511.
SIAM, 2007. 42
[Ach07]
Tobias
Achterberg.
Constraint
Integer
Programming.
PhD
thesis,
Technische
Universitat
Berlin,
2007.
http://opus.kobv.de/tuberlin/volltexte/2007/1611/. 117, 118, 139
[AF05]
[AKPW95] Noga Alon, Richard M. Karp, David Peleg, and Douglas B. West. A graphtheoretic game and its application to the k-server problem. SIAM J. Comput.,
24(1):78100, 1995. 3, 42, 43, 45, 46, 77, 78, 79, 84, 87, 92
[All68]
[ALMM04] Edoardo Amaldi, Leo Liberti, Nelson Maculan, and Francesco Maffioli. Efficient edge-swapping heuristics for finding minimum fundamental cycle bases. In
Celso C. Ribeiro and Simone L. Martins, editors, WEA, volume 3059 of Lecture
Notes in Computer Science, pages 1429. Springer, 2004. 42, 43, 44, 45, 57, 60,
70, 85, 86, 88
[Ant75]
Jean Antoniadis. M
oglichkeiten des Einsatzes des Operations Research, insbesondere der linearen Programmierung und ihrer Erweiterungen, zur Optimierung
der Steuerung von Lichtsignalanlagen. PhD thesis, TU Berlin, 1975. in German.
10
[BDL+ 03]
Andreas Brandst
adt, Feodor F. Dragan, Ho`ang-Oanh Le, Van Bang Le, and
Ryuhei Uehara. Tree spanners for bipartite graphs and probe interval graphs.
In Hans L. Bodlaender, editor, WG, volume 2880 of Lecture Notes in Computer
Science, pages 106118. Springer, 2003. 110
143
144
Bibliography
[BDLL04]
Andreas Brandst
adt, Feodor F. Dragan, Ho`ang-Oanh Le, and Van Bang Le.
Tree spanners on chordal graphs: complexity and algorithms. Theor. Comput.
Sci., 310(1-3):329354, 2004. 110
[BE05]
Simone B
achle and Falk Ebert. Graph theoretical algorithms for index reduction
in circuit simulation. Preprint 04-245 (Matheon), TU Berlin, Mathematical
Institute, 2005. 42
[Ber62]
[BKW03]
Philipp Boksberger, Fabian Kuhn, and Roger Wattenhofer. On the approximation of the minimum maximum stretch tree problem. Technical Report 409,
2003. http://www.inf.ethz.ch/research/disstechreps/techreports. 77,
80
[Bod93]
Hans L. Bodlaender. A tourist guide through treewidth. Acta Cybern., 11(12):122, 1993. 43
[Bok03]
[Bol02]
Bela Bollobas. Modern Graph Theory, volume 184 of Graduate Texts in Mathematics. Springer, 2002. 2nd printing. 41, 123
[BW05]
Robert Braun and Florian Weichenmeier. Automatic offline-optimization of coordinated traffic signal control in urban networks using genetic algorithms. In
Proceedings of the 12th World Congress on Intelligent Transport Systems, San
Francisco, 06.-10. Nov. 2005, 2005. 11
[CC95]
Leizhen Cai and Derek G. Corneil. Tree spanners. SIAM J. Discrete Math.,
8(3):359387, 1995. 3, 92, 94, 97, 110
[CGH95]
[CI88]
Guilio Erberto Cantarella and Gennaro Improta. Capacity factor or cycle time
optimization for signalized junctions: A graph theory approach. Transportation
Research B, 22B(1):123, 1988. 11
[CPL07]
[Dag95]
Carlos F. Daganzo. The cell transmission model, part ii: Network traffic. Transportation Research B, 29(2):7993, 1995. 11
John Wi-
Diploma thesis,
ILOG CPLEX
[DDGS03] Elias Dahlhaus, Peter Dankelmann, Wayne Goddard, and Henda C. Swart.
MAD trees and distance-hereditary graphs. Discrete Applied Mathematics,
131(1):151167, 2003. 92
[Die00]
[DK08]
[DKP82]
Narsingh Deo, M.S. Krishnomoorthy, and G.M. Prabhu. Algorithms for generating fundamental cycles in a graph. ACM Transactions on Mathematical
Software, 8(1):2642, 1982. 3, 4, 42, 92, 94, 110
Bibliography
145
[DKP95]
Narsingh Deo, Nishit Kumar, and James Parsons. Minimum-length fundamental-cycle set problem: A new heuristic and an SIMD implementation. Technical
Report CS-TR-95-04, University of Central Florida, Orlando, 1995. 42, 43, 87
[DMN85]
[EEST05]
Michael Elkin, Yuval Emek, Daniel A. Spielman, and Shang-Hua Teng. Lowerstretch spanning trees. In Harold N. Gabow and Ronald Fagin, editors, STOC,
pages 494503. ACM, 2005. 42, 92, 94, 114
[ELR07]
Michael Elkin, Christian Liebchen, and Romeo Rizzi. New length bounds for
cycle bases. Inf. Process. Lett., 104(5):186193, 2007. 42
[EP04]
[FB93]
[FK01]
S
andor P. Fekete and Jana Kremer. Tree spanners in planar graphs. Discrete
Applied Mathematics, 108(1-2):85103, 2001. 110
[GA03]
[Gal01]
Giulia Galbiati. On min-max cycle bases. In Peter Eades and Tadao Takaoka,
editors, ISAAC, volume 2223 of Lecture Notes in Computer Science, pages 116
123. Springer, 2001. 91, 93, 114, 115
[Gal03]
Giulia Galbiati. On finding cycle bases and fundamental cycle bases with a
shortest maximal cycle. Inf. Process. Lett., 88(4):155159, 2003. 91, 93, 114,
115
[Gar72]
[GC67]
Calvin C. Gotlieb and Derek G. Corneil. Algorithms for finding a fundamental set of cycles for an undirected linear graph. Communications of the ACM,
10(12):780783, 1967. 42
[Ger05]
[GJ79]
[Gle01]
Petra M. Gleiss. Short Cycles. PhD thesis, Universitat Wien, 2001. 41, 123
146
Bibliography
[GLG75a]
[GLG75b]
[GLG76]
[GRA08]
[Hae04]
[Has96]
Refael Hassin. A flow algorithm for network synchronization. Operations Research, 44(4):570579, 1996. 11, 15, 17, 18
[hcm00]
[Hil65]
John A. Hillier. Glasgows experiment in area traffic control. Traffic Engineering & Control, 7(8):502509, 1965. 9
[Hil66]
[HL02]
Refael Hassin and Asaf Levin. Minimum restricted diameter spanning trees.
In Klaus Jansen, Stefano Leonardi, and Vijay V. Vazirani, editors, APPROX,
volume 2462 of Lecture Notes in Computer Science, pages 175184. Springer,
2002. 4, 92, 94
[HO02]
Lane A. Hemaspaandra and Mitsunori Ogihara. The complexity theory companion. Springer-Verlag New York, Inc., New York, NY, USA, 2002. 6
[Hor87]
[HR67]
John A. Hillier and Richard Rothery. The synchronization of traffic signals for
minimum delay. Transportation Science, 1:8194, 1967. 9
[HT95]
Refael Hassin and Arie Tamir. On the minimum diameter spanning tree problem.
Inf. Process. Lett., 53(2):109111, 1995. 110
[Hu74]
[Ian94]
Stefano Ianigro.
Ein Modell zur Simulation eines innerst
adtischen
Verkehrsablaufes und zur Steuerung von Lichtsignalanlagen mittels Petri Netzen unter Ber
ucksichtigung der Gr
unen Welle. PhD thesis, Universitat der
Bundeswehr Hamburg, 1994. 11
[IS82]
Gennaro Improta and Antonio Sforza. Optimal offsets for traffic signal systems
in urban networks. Transportation Research Board, 16B(2):143161, 1982. 10
Bibliography
147
[JLK78]
David S. Johnson, Jan Karel Lenstra, and Alexander H.G. Rinnoy Kan. The
complexity of the network design problem. Networks, 8:279285, 1978. 92, 94,
110, 111
[Kir47]
Ekkehard K
ohler, Rolf H. Mohring, and Gregor W
unsch. Minimizing total delay
in fixed-time controlled traffic networks. In Hein Fleuren, Dick den Hertog,
and Peter Kort, editors, Operations Research Proceedings 2004, pages 192199.
Springer, 2005. 41
[KN03]
Jitka Kratochvilova and Ivan Nagy. Bibliographic search for optimization methods of signal traffic control. Technical Report 2081, Institute of Information
Theory and Automation, Academy of Sciences of the Czech Republic, Prague,
2003. 11
[LAM05]
Leo Liberti, Edoardo Amaldi, and Francesco Maffioli Nelson Maculan. Mathematical models and a constructive heuristic for finding minimum fundamental
cycle bases. Yugoslav Journal of Operations Research, 15(1):1524, 2005. 42,
43, 44, 57, 60, 61, 87
[Lam07]
Stefan L
ammer. Reglerentwurf zur dezentralen Online-Steuerung von Lichtsignalanlagen in Straennetzwerken. PhD thesis, TU Dresden, 2007. In German.
7
[Lie05]
[Lie06]
Christian Liebchen. Periodic Timetable Optimization in Public Transport. dissertation.de Verlag im Internet, Berlin, 2006. 17, 26, 27, 41, 42, 123, 124,
128
[Lit66]
[LP02]
Christian Liebchen and Leon Peeters. On cyclic timetabling and cycles in graphs.
Preprint 761, TU Berlin, Mathematical Institute, 2002. 26
148
Bibliography
[LPW05]
Christian Liebchen, Mark Proksch, and Frank H. Wagner. Performance of algorithms for periodic timetable optimization. In Mark Hickman, editor, ComputerAided Transit Scheduling Proceedings of the Ninth International Workshop
on Computer-Aided Scheduling of Public Transport (CASPT), Lecture Notes in
Economics and Mathematical Systems. Springer, 2005. To appear. 27
[LR07]
Christian Liebchen and Romeo Rizzi. Classes of cycle bases. Discrete Applied
Mathematics, 155(3):337355, 2007. 26, 44
[LW07]
John T. Morgan and John D. C. Little. Synchronizing traffic signals for maximal
bandwidth. Operations Research, 12:897912, 1964. 9
[MNW06]
Rolf H. M
ohring, Klaus Nokel, and Gregor W
unsch. A model and fast optimization method for signal coordination in a network. In Proceedings of the 11th
IFAC Symposium on Control in Transportation Systems, pages 7378, Delft,
The Netherlands, August 2006. 7
[MOP07]
MOPS
Mathematical
OPtimization
http://www.mops.fu-berlin.de/, 8.0 edition, 2007. 117, 118, 139
[Nac96a]
Karl Nachtigall. Cutting planes for a polyhedron associated with a periodic network. Preprint IB 112-96/17, Deutsche Forschungsanstalt f
ur Luft und Raumfahrt, 1996. 19
[Nac96b]
[net80]
Traffic network analysis with NETSIMA user guide, 1980. Federal Highway
Administration, Washington D.C. 129
[New64]
[NW88]
George L. Nemhauser and Laurence A. Wolsey. Integer and Combinatorial Optimization. Wiley, New York, 1988. 6
[Nok06]
Klaus N
okel. Personal Communication, 2006. 136
[Odi94]
[Odi96]
[Pat69]
[PR99]
David Peleg and Eilon Reshef. A variant of the arrow distributed directory
with low average complexity. In Jir Wiedermann, Peter van Emde Boas, and
Mogens Nielsen, editors, ICALP, volume 1644 of Lecture Notes in Computer
Science, pages 615624. Springer, 1999. 93, 114
System.
149
Bibliography
[PT01]
David Peleg and Dov Tendler. Low stretch spanning trees for planar graphs.
Technical Report MCS01-14, Weizmann Institute of Science, 2001. 4, 92
[ptv]
[RB91]
[ril92]
Richtlinien f
ur Lichtsignalanlagen RiLSA, Lichtzeichenanlagen f
ur den Straenverkehr, 1992. Forschungsgesellschaft f
ur Straen- und Verkehrswesen, Koln,
Germany. 6, 8
[Rob69]
[RS86]
[RS90]
Ajay K. Rathi and Alberto J. Santiago. The new NETSIM simulation program.
Traffic Engineering and Control, 31(5):317320, 1990. 129
[Sch86]
[Sch03]
[SS95]
Ziad A. Sabra and Charles R. Stockfisch. Advanced traffic models: State of the
art. ITE Journal, pages 3142, 1995. 7
[Sto68]
[SU89a]
Paolo Serafini and Walter Ukovich. A mathematical model for periodic scheduling problems. SIAM Journal on Discrete Mathematics, 2(4):550581, 1989. 11,
16, 18, 36
[SU89b]
Paolo Serafini and Walter Ukovich. A mathematical model for the fixed time
traffic control problem. European Journal of Operations Research, 42:152165,
1989. 11, 16, 17, 36
[Suh07]
[syn00]
SYNCHRO, Users Guide, 2000. Trafficware, Sugar Land, USA. 12, 13, 37
[tft]
Traffic
flow
theory,
a
state
http://www.tfhrc.gov/its/tft/tft.htm. 7
[WCT00]
[Web58]
F.V. Webster. Traffic signal settings. Technical Report 39, Her majestys stationery office, 1958. Road Research Technical Paper. 9
[Wes96]
[Whi35]
of
the
art
report.
150
Bibliography
[WLB+ 99] Bang Ye Wu, Giuseppe Lancia, Vineet Bafna, Kun-Mao Chao, R. Ravi, and
Chuan Yi Tang. A polynomial-time approximation scheme for minimum routing
cost spanning trees. SIAM J. Comput., 29(3):761778, 1999. 92
[Wor65]
R. Wormleighton. Queues at a fixed time traffic signal with periodic random input. Canadian Operational Research Society (CORS) Journal, 3:129141, 1965.
20
[Wun96]
Roland Wunderling.
Paralleler und objektorientierter SimplexAlgorithmus.
PhD thesis, Technische Universitat Berlin, 1996.
http://www.zib.de/Publications/abstracts/TR-96-09/. 118