 Inventory Strategy Transport Strategy

 Forecasting
 Storage fundamentals  Transport fundamentals
 Inventory decisions
 Purchasing and supply  Transport decisions
scheduling decisions
 Storage decisions

Customer Service
Goals

Location Strategy
 Location decisions
 The network planning process

Importance of an Effective Transportation

System
• Accessibility to markets
• Greater competition
– more distant markets can be served
• Economies of scale
– wider markets => greater production volume
– production points need not be close to markets
• Lower prices
– increased competition among suppliers
– lower production and transportation costs

1
Basic Transportation Modes
• Rail
• Truck
• Air
• Water
• Pipeline

Rail
• long haul
• slow mover of raw materials
• large loads
• common carrier or private
• carload, less-than-carload, multiple carload
• consolidation, stop-off, re-route

Costs
• high fixed costs
– terminal costs, loading/unloading, yard switching of multi-
product multi-shipment trains
• low variable costs
– proportional to distance

2
 Truck
• Medium haul:
• Medium Load
• Door-to-door convenience
• good speed and frequency (small dispatch lots)
• cannot carry large loads
• common, contract or private
Costs
• Low fixed costs
– carrier do not own roads
• high variable costs (50-60 %)
– fuel, tolls, taxes, maintenance

Air
• expensive (2  truck, 16  rail)
• avg. haul = 1,300 miles
• fastest speed
– 565 mph air travel speed
– terminal time (taxi-ing, holding, ground handling)
• constrained by cargo space and lifting capabilities
• security high
Costs
• Fixed costs
– terminal and equipment depreciation
• Variable costs
– inversely related to length of haul
• Total cost high, esp. for short deliveries

3
Pipeline
• limited capabilities
• crude oil, water, Milk ?
• slow
• 24-hour service
• high capacity
• Reliable and low risk of disruption and damage
Costs
• high fixed costs
– pipes, pumping equipment, own or lease right-of-way
• Variable costs
– pump operation, depends on throughput and pipe diameter
– loss through seepage

• India recently completed one of its most

ambitious projects, the 1,700 km Hazira-
Bijaipur-Jagdishpur pipeline. Costing nearly
Rs. 17 billion, the pipeline transports liquid
gas from the South Bassein offshore field off
Mumbai to Jagdishpur and Aonla, deep in the
mainland in Uttar Pradesh. Besides, India has
nearly 7,000 km of pipeline mainly for the
transportation of crude oil and its products

4
Water
• limited in scope
– inland waterways, coastal
• heavy , bulk commodities
• slow
• affected by weather (freezing, floods)
Costs
• Fixed costs
– mainly transport equipment
– waterways and harbours publicly owned
– terminal costs: harbour fees, loading/unloading(high costs if
not containerised)
• Variable costs (low)
– no charge for use of waterways
• favours bulk commodity goods

Intermodal Transport
• Truck-Rail “piggyback”
• Truck-water “fishyback”

Trailer on Flat Car (TOFC)

• long haul cost economy of rail
• convenience and accessibility of trucks at
origin/destination
• shipper: door-to-door service at lower than truck rates
• rail: more business
e.g.Ro-Ro Scheme of Konkan Railways

5
Types of Inter-modal Options

How do you call this?

6
Roll On -Roll Off
• Konkan Railway Corporation Introduced
a special service for handling cargo
from Verna (Goa) to Kallad( Near
Mumbai).
• The service is Roll on - Roll Off
• A Rail -Road Synergy.

7
C|/Program Files/Netscape/Users/tasviji/mail/Inbox?id=B39174A1004FD211A2E000A0C9DCE7D622037C%40ihqnt001.hq.m

What is RO-RO?
• Ro-Ro stands for roll on roll off.
• Trucks are carried in specially designed
cabins and hauled along with other
wagons.

8
Ro - Ro Redefines Logistics.
• Features of Ro-Ro
– Faster Reach.
– Safer.
– Reliable.
– Economical (For both truck owners and
Mfgs)
– Freight on truck basis.
– Extended life of trucks (less wear &tear)

Ro - Ro Redefines Logistics.
• No Hassles of Toll Gate.
• High Cargo Velocity.
• Freight Rs 4500/per truck load.
• No re handling at railway station, no
losses no pilferage.
• Normal Freight is Rs 675-00 per ton on
road basis.
• For Ro-Ro it is Rs 525-00 per ton

9
Traffic
• Every day 32 truck take off from Verna.
• More trips if there is more demand.
• Customer Response: Transport agents
are very happy about it. Manufacturer’s
are also happy.( Freight /ton is reduced
by Rs 150)

Creativity of Konkan Railways!!!

This is Rail road!!! What about Road Trains? No jokes….

10
11
12
Containerised Freight (COFC)

• first trip: trailers on a WWII tanker

from New Jersey to Texas in 1956
• soon after: specially converted ships to
stack van-sized boxes on deck
• standard size avoids re-handling
– 8 x 8 x 20 (TEU)
– 8 x 8 x 40 or 8 x 8 x 45

Transportation Costs
• Fixed Costs
– road/railway acquisition and maintenance,
terminal facilities, transport equipment, carrier
administration
• Variable Costs
– fuel, labour, equipment maintenance, handling,
pickup and delivery
• Cost Allocation Difficult
– By shipment? Weight? Volume?
– Insurance value? Delivery guarantees?

Back Haul Costs?

13
Service Choices
• Agents
– handle small shipments
– consolidate them into vehicle load quantities
– own little or no line-haul equipment
– pick-up / delivery service to shippers
1. Air and surface freight forwarders
– for-hire carriers of freight
– purchase long-distance service from air, truck, rail
and water carriers

Service Choices
• Agents
2. Shippers associations
3. Transport brokers
– bring shippers and carriers together
– provide information about rates, routes, capabilities
– arrange transportation

• Small shipment services

14
 The characteristics of each transport mode may make that
mode more or less attractive depending upon the relevant
product and customer service characteristics.

Comparisons of alternative transport choices

Size and Weight Restrictions
Air: Little or no flexibility
Sea: Virtually Unlimited
Rail: Some dimension restrictions(height) none for
weight
Road: Height and Weight restrictions
Safety

Air Highly regulated

Sea Less regulated than air
Rail Few restrictions
Road Less regulated than air

Fragility
Air Reliable product handling
Sea Variable reliability
Rail Transport vehicles can have vibration and shunting
problems
Road Handling problems can occur often

Temperature
Air Temperature controlled containers are available
Sea Controlled
Rail Controlled
Road Controlled
Cost per unit
Air High
Sea Low
Rail Low
Road Moderate, can vary with service requirements

15
 Transit time and reliability

Air Fast and reliable

Sea slow reliability varies :full container loads

(fcls) are more reliable
Rail Dedicated services reliable part shipments have
difficulties
Road Versatile and usually offer good transit times;
loading and unloading can slow progress.
Reliability varies with the service policy of each
company.

Design options

• Direct • Milk runs

shipment

Suppliers Retail
stores
Suppliers Retail Suppliers Retail
stores stores

16
 Design options
• Using DCs

DC
DC

Suppliers Retail Suppliers

stores
Retail
stores

Network structure Pros Cons

Direct shipping
Direct shipping with milk
runs
All shipments via central DC
with inventory storage
•no intermediate network
•simple to coordinate
•lower transportation costs
for small lots
•lower inventories
•lower inbound
transportation cost through
consolidation
•high inventories (due to
large lot size)
•significant receiving
expense
•increased coordination
complexity
•increased inventory cost
•increased handling at DC

All shipments via central DC •very low inventory •increased coordination

with crossdocking
Shipping via DC using milk
runs
requirement complexity
•lower transportation cost
through consolidation
•lower outbound •further increase in
transportation cost for small coordination complexity
lots

Tailored network * transportation choice best •highest coordination

matches needs of individual complexity
product and customer

17
 Service company selection :some considerations

Should carriers feature in the logistics activity some criteria

should be established for their selection. The carrier selection
process is complex and should begin with identifying carrier
attributes required to implement the company’s customer
service policy.

They should be rank ordered to reflect importance. Potential

service companies should then evaluated against these criteria
to determine those most suitable.

Among the many attributes used, the following represent a typical

spread from which an effective evaluation may be determined

Current performance : delivery accuracy and reliability

Acceptance of responsibility : a willingness to guarantee the service
product offered.
Flexibility: an ability to respond to emergency situations.
Comprehensive services offer : the range of services offered, their
capacity and the company’s capabilities.
Information systems; information availability is essential for control
purposes.
Financial stability: is necessary to ensure continuity of service and
for the upgrade of the equipment within the business.
Compatibility : the service company’s management should share
the same, or similar, views on objectives and strategy for their
business. They should be capable of integrating into the customer
business, sharing its confidences and its problems.

18
NODES VS LINKS PERSPECTIVE OF LOGISTICS
ENABLES TO USE OR/MS APROACH TO LOGISTICS

Use of Quantitative OR/MS models in Logistics

Various quantitative models from the Operations

Research/Management Science (OR/MS) are of
immense help in many logistics decisions.

 Vehicle operating cost models

 Composite models
 Location models
 Allocation models
 Network Models
 Inventory models
 Routing models

and many more

19
 VEHICLE ROUTING AND SCHEDULING
Principles for Good Routing and Scheduling
1. Load trucks with stop volumes that are in the closest proximity to each other

2. Stops on different days should be arranged to produce tight clusters.

3. Build routes beginning with the farthest stop from the depot.

4. The sequence of stops on a truck route should form a teardrop pattern.

5. The most efficient routes are built using the largest vehicles available.

6. Pickups should be mixed into delivery routes rather than assigned to the end
of routes.

7. A stop that is greatly removed from a route cluster is a good candidate for an
alternate means of delivery.

8. Narrow stop time window restrictions should be avoided.

Methods for Routing and Scheduling

 Network Analysis

The Sweep Method

The Savings Method

 Route Sequencing

20
 NETWORK MODELS

Consider the following situations

 Design of off shore natural-gas pipeline network connecting

wellheads with an onshore delivery point with the objective of
minimizing the cost of constructing the pipeline.

 Determination of the shortest route joining two cities in an

existing network of roads.

 Determination of the maximum annual capacity in tons of a coal

slurry pipeline network joining the coal mines with the power
plants (Slurry pipelines transport coal by pumping water through
suitably designed pipes operating between the coal mines and the
desired destinations.)

 Determination of the minimum-cost flow schedule from oil

fields to refineries and finally to distribution centers.
Crude oils and gasoline products can be shipped via tankers,
pipelines, and/pr trucks. In addition to maximum supply availability
at the oil fields and minimum demand requirements at the
distribution centers, restrictions on the capacity of the refineries and
the modes of transportation must be taken into account

The list reveals that network optimization problems can generally be

modeled by one of four models:

1. Minimal spanning tree model (situation 1 )

2.Shortest-route model (situation 2)
3.Maximum-flow model (situation 3)
4.Minimum-cost capacitated network model (situation 4)

21
From the practical standpoint, the minimum-cost capacitated
network model enjoys a wide variety of applications. Indeed,
both the shortest-route and maximum-flow problem can be
formulated as special cases of the capacitated transportation
model.

NETWORK DEFINITIONS
 A network consists of a set of nodes linked by arcs or branches.
Associated with each arc is a flow of some type. For example,
in a transportation network, cities represent nodes and highways
represent arcs, with traffic representing the arc flow

 The standard notation for describing a network G is G=(N,A)

where N is the set of nodes and A is the set of arcs.

The network below is comprised of five nodes and eight arcs

1 5
3

2 4

N={1,2,3,4,5}
A={(1,3,), (1,2), (2,3), (2,4), (2,5), (3,4), (3,5), (4,5)}

22
Associated with each network is a flow of some type (e.g. oil
products flow in pipeline network and network and traffic flows
in a transportation network.

 In general, the flow in an arc is limited by its capacity, which

may be finite or infinite. An arc is said to be directed or oriented
if it allows positive flow in one direction and zero flows in the
opposite direction.

 A directed network is a network with all directed arcs.

 A path is a sequence of distinct arcs that join the two nodes

regardless of the orientation of the individual arcs. For example
in the figure, arcs (1,3), (3,2), and (2,4) represent a path from
node 1 to node 4.

 A path will form a loop or a cycle if it connects a node to

itself. For example , in Figure the arcs (2,3), (3,4), and (4,2)
form a loop.

 A directed loop (or a circuit) is a loop in which all the arcs

have the same direction or orientation.

 A connected network is network in which every two distinct

nodes are linked by a path as demonstrated by the network in
the Figure.

 A tree is a connected network that may involve only a subset

of the nodes
 A spanning tree is a connected network that includes all the
nodes in the network with no loops.

23
 The following defines a tree for the network in the earlier
Figure.
1 3

2 4
Tree

The following defines a spanning tree for the network in the

earlier Figure

1 3 5

4
2

24
 MINIMAL SPANNING TREE PROBLEM

Consider the situation in which it is desired to create a network of

paved roads that link a number of rural towns. Because of budget
limitations, the miles of constructed roads must be the absolute
minimum that allows direct or indirect traffic linkage among the
different towns.

The situation above can be represented by a network in which towns

represent nodes and proposed roads represent arcs. The resulting
model is typical of the so called minimal spanning tree problem,
where it is desired to determine the spanning tree that results in the
smallest sum of the connecting arcs. In effect, the minimal
spanning tree deals with finding the most “efficient” connections
among all the nodes in the network, which, by definition, cannot
include any loops or cycles.

Minimal Spanning Tree Algorithm

 The minimal spanning tree algorithm calls for starting with any
node and joining it to the closest node in the network.

 The resulting two nodes then form a connected set, C, with the
remaining nodes comprising the unconnected set, .

 Next, we choose a node from the unconnected set that is closest

(has the shortest arc length) to any node in the connected set. The
chosen node is then removed from the unconnected set and joined to
the connected set.

 The process is repeated until the unconnected set becomes empty

(or equivalently, until all the nodes are moved from  to set C.) A
tie be broken arbitrarily. Ties, however, point to the existence of
alternative solutions.

25
A TV cable Company is in the process of planning a network for
providing cable TV service to five new housing development areas.
The cable system network is summarized below
3 (mil es)
2 5
1
4 6
9
1

5 3 8
10

7 5
6

4 3

The numbers associated with each branch represent the miles of

cable needed to connect any two locations. Node 1 represents the
cable TV relay station and the remaining nodes (2 through
6)represent the five development areas. A missing branch between
two nodes implies that it is prohibitively expensive or physically
impossible to connect the associated development areas. It is
required to determine the links that will result in the use of
minimum cable miles while guaranteeing that all areas are
connected (directly or indirectly) to the cable TV station.

The graphical solution is summarized in Figures by iterations. The

procedure can be started from any node, always ending up with the
same optimum solution. In the cable TV example, it is logical to
start the computations with node 1. The corresponding set of
“unconnected nodes” is represented by nodes 2,3,4, 5 and 6.
Symbolically, we write This as C={1},  ={2, 3, 4, 5, 6}

26
Iteration 1

Node 1 must be connected to node 2, the closest node in C={2, 3,

4, 5, 6}. Iteration 1 of Figure thus shows that C={1,2},  ={3, 4,
5, 6}

Iteration 2

Nodes 1 and 2 {set C} are now linked permanently. In iteration 2

we select a node in  ={3, 4, 5, 6} that is closest to a node in
C={1,2}. Since the shortest distance occurs between nodes 2 and 5
(see iteration 2 of Figure ), we have C={1, 2, 5},  ={3, 4, 6}

Connected set C

3
2
Unconnected set 
6
4
1 5
9

1 5
3

7
6
4
Iteration 2

27
Iteration 3
Iteration 3 of Figure gives the distance from the nodes of
{1, 2, 5} to all the nodes in  =(3, 4, 6}. Thus nodes 2 and 4 are
connected, which yields C={1, 2, 4, 5}, ={3, 6}

C
3
2
6 5
1

3
1
5

4
7
8 6

4


28
Iteration 4
Iteration 4 of Figure shows that nodes 4 and 6 must be
connected. Thus we obtain C={1, 2, 4, 5, 6},  ={3}

3
2
5
1 6
4

1
5
3
6
5 

4
3
Iteration 4

29
 Iteration 5
In iteration 5 we have a tie, that may be broken
arbitrarily. This means that we can connect 1 and 3 or 4
and 3. Both (altenative) solutions lead to

3 C
2
6
4
5
1

1
5  Iteration 5
3

6
4 3

30
 C={1, 2, 3, 4, 5, 6},  =0
Since all the nodes are connected, the procedure is complete. The
minimum cable miles that are used to connect the development areas
to the TV station equal 1 + 3 + 4 + 3 + 5 = 16 miles.

2 3

5
1 4

1
5
3
6
5
Alternate
links 4
3
Minimal spanning tree

31
ORIGIN
B E
Amarillo 90 minutes 84 84 I
A

66 120
138 132

126
C F
90 H
60

348 132
156
126
48

48
J
D 150
G
DESTINATION

Step Solved nodes Its closest connected Total nth Its Its
Directly connected unsolved Cost nearest minimum last
to unsolved nodes nodes Involved node cost connection

1 A B 90 B 90 AB*

2 A C 138 C 138 AC
B C (90+66) 156

3 A D 348
B E (90+84) 174 E 174 BE*
C F (138+90) 228

4 A D 348
C F (138+90) 228 F 228 CF
E I (174+84) 258

5 A D 348
C D (138+156) 294
E I (174+84) 258 I 258 EI*
F H (228+60) 288
6 A D 348
C D (138+!56) 294
F H (228+60) 288 H 288 FH
I J (258+126) 384

32
 7 A D 348
C D (138+156) 294 D 294 CD
F G (288+132) 360
H G (288+48) 336
I J (258+126) 384

8 H J (288+126) 414
I J (258+126) 384 J 384 IJ*

A Asterisk (*) denotes minimum cost route.

A-B-E-I-J

Maximum Flow Problem

A 3
0 0
1 T
1 0 9
0
0 B 4 0 D
5 5 1
2
O 7
4
1
0
0 6
E
C 4 0
0

33
  At the outset, the residual network differs from the original
network only in that each directed are (i j) lacking a directed are
in the opposite direction (j i) now has such an arc added with
zero arc capacity.

 Subsequently, the arc capacities in the residual network (called

residual capacities) are adjusted as follows. Each time some
amount of flow is added to arc i j in the original network, the
residual capacity of arc i j is decreased by but the residual
capacity of arc j i is increased by

 Thus the residual capacity represents the unused arc capacity in

the original network or the amount of flow in the opposite
direction in this network that can be cancelled ( or a combination
of both if the original network has arcs in both directions).

 Therefore, after assigning various flows to the original network,

the residual network shows how much more can be done either by
increasing flows further or by canceling previously assigned flows

 An augmenting path is a directed path from the supply node to

the demand mode in the residual network such that every arc on this
path has strictly positive residual capacity.

 The minimum of these residual capacities is called the residual

capacity of the augmenting path because it represents the amount of
flow that can feasibly be added to the entire path.

 Therefore, each augmenting path provides an opportunity to

further augment the flow through the original network.

34
  The augmenting path algorithm repeatedly selects some
augmenting path and adds flow equal to its residual capacity to
that path in the original network.

 This process continues until there are no more augmenting

paths, so the flow form the supply node to the demand node
cannot be increased further.

 The key to ensuring that the final solution necessarily is

optimal is the fact that augmenting paths can cancel some
previously assigned flows in the original network, so an
indiscriminate selection of paths for assigning flows cannot
prevent the use of a better combination of flow assignments.

Algorithm for Maximum Flow Problem

1.Identify an augmenting path by finding some directed path from
the supply node to the demand node in the residual network such that
every arc on this path has strictly positive residual capacity. (If no
augmenting path exists;the net flows already assigned constitute an
optimal flow pattern.)

2.Identify the residual capacity c* of this augmenting path by

finding the minimum of the residual capacities of the arcs on This
path. Increase the flow in This path by c*.

3.Decrease by c* the residual capacity arc on this augmenting path.

Increase by c* the residual capacity of each arc in the opposite
direction on this augmenting path. Return to step 1.

35
 Iteration 1: Referring to Fig . One of several augmenting paths is
O B E T, which has a residual capacity of min {7,5,6}=5.Assigning
a flow of 5 to This path, the resulting residual network is

A 3 0
T 5
1 5
0
0 9
0 D
1
1 4
5 B 0
2
5 O 2
4

0 1
0 C 5 1
4 E
0

Iteration 2: Assign a flow of 3 to the augmenting path , 0 A D T .

The resulting residual network is

A 0 8
3 T
1 5
3 3 6
0 D 1
2
1 4
5
B 0
2 2
8 O
4

0 1
5
0 C 1
4 E
0

36
 Iteration 3:Assign a flow of 1 to the augmenting path ,O -A-B-D-T.
Iteration 4: Assign a flow of 2 to te augmenting path, O-B-D-T. The
resulting residual network is
A 0 6 T 11
0
5
4 3
3
3 D
1
2
1 7 1
B 0
0
O 2
11
4

0 1
0 C 5 1
4 E
0

Iteration 5: Assign a flow of 1 to the augmenting path, O-C-E-D-T..

Iteration 6: Assign a flow of 1 to the augmenting path , O-C-E-T. The
resulting residual network is

A 0 7
0 T 13
4 3 2
3 D 6
1 2
2
7 1
B 0
13 0 2
O
2
0
0 2 0
2 C 5
2 E

37
 Iteration 7: Assign a flow of 1 to the augmenting path, O-C-E-B-D-T.
The resulting residual network is

4 A 0 8 T 14
0 3 6
1
4 D
2
1 2
7 0
B 1
0
14 O 2
1

0 0
4
3 C 3 0
1 E

14
T
A
3 8

D
1
4
4
7 B 6
14 O 1

4
3

C
E
3

38
 0 7 T 14
A 0 6
4 3
2
1 3 D
2
1
1 7 B 0
14 O 0 2
2

0 0
0
2 C 5
2 E
2

3 0T
A 1 0
0 0
9
D
1 0

1 4
5 0
B 5
O 7 2
4

0 1
0 6
0 C
4 0 E

39
 Vehicle Routing Decisions

Vehicle Routing
• Reduce transportation cost and improve
customer service level
• Find the best path that a vehicle should
follow through a network of roads, rail
lines, routes that will minimize time or
distance

40
Shortest Path Problem
• Solution
– Using the Solver
– Special algorithms
• The “costs” (on the links, or arcs)
– transportation cost per unit distance traveled
– minimize total transportation cost (find least costly path)
– travel time per unit distance
– minimize total travel time (find shortest path)
• The “links” (or “arcs”)
– may be provided by more than one transport
provider
(multiple modes of transport also possible)

Poor routing-paths cross Good routing-no paths cross

D D
Depot
Depot

41
Stops

D
D

42
 T T

F T F

T T

F
F

F
STOP

Depot

T
F
T

F F
T
F

T
T
F
F

T
T
F

Depot

43
Heuristic - Sweep Method

vehicle capacity = 5000 units

2000
2000
2000
2000
1000
2000
2000 Sweep method:
2000 - start in any direction
1000
1000 - move clockwise
2000 1000 - a route is defined when
1000 cumulative capacity
of stops exceed vehicle
capacity

Heuristic - Savings Method

• Objective: Minimize total distance
traveled by all vehicles
• Compute maximum distance traveled by
assuming each stop is served by one
vehicle
• Look at pairs of stops and combine the
pair with largest “savings” in a route

44
Example - Savings Method

A 3000 Distances
W-A: 40
C 1000 W-C: 50
B W-B: 90
W-D: 70
W 3000 A-C: 35
A-B: 80
A-D: 85
D B-C: 50
1000 B-D: 80
C-D: 50
vehicle capacity = 5000 units solution: 2 vehicles, (A), (C,B,D)

Shortest Path Problem

• A network of links and terminals
• There is a cost (or time or “weight”)
associated with each link
• Find the shortest path that connects the
origin to the destination on the network
• Solution ?

45
 Shortest Path Problem
• Single origin, single destination

D
B 348
A 90 60
149 F
138 C 93
origin
(source) destination
150 100 (sink)

cost or time E

Shortest Path Problem - Formulation

46
 B D
348
A 90 60
149 F
138 C 93

150 100
E

USING THE SOLVER FOR THE SHORTEST PATH PROBLEM

47
48