Download as pdf or txt
Download as pdf or txt
You are on page 1of 31

sustainability

Article
Electric Vehicle Charging Station Location-Routing Problem
with Time Windows and Resource Sharing
Yong Wang 1, * , Jingxin Zhou 1 , Yaoyao Sun 2 , Xiuwen Wang 3 , Jiayi Zhe 1 and Haizhong Wang 4

1 School of Economics and Management, Chongqing Jiaotong University, Chongqing 400074, China
2 School of Economics and Management, Nanjing University of Aeronautics and Astronautics,
Nanjing 211106, China
3 School of Management, Shanghai University, Shanghai 200444, China
4 School of Civil and Construction Engineering, Oregon State University, Corvallis, OR 97330, USA
* Correspondence: yongwx@cqjtu.edu.cn

Abstract: Electric vehicles (EVs) are widely applied in logistics companies’ urban logistics distri-
bution, as fuel prices increase and environmental awareness grows. This study introduces an EV
charging station (CS) location-routing problem with time windows and resource sharing (EVCS-
LRPTWRS). Resource sharing, among multiple depots within multiple service periods is proposed
to adjust the transportation resource configuration for a sustainable logistics development. Solving
the EVCS-LRPTWRS involves a periodic CS location selection and a multi-depot multi-period EV
routing optimization. A bi-objective nonlinear programming model is proposed to formulate the
EVCS-LRPTWRS with a minimum total operating cost and number of EVs. A hybrid algorithm
combining the Gaussian mixture clustering algorithm (GMCA) with the improved nondominated
sorting genetic algorithm-II (INSGA-II) is designed to address the EVCS-LRPTWRS. The GMCA is
employed to assign customers to appropriate depots in various service periods in order to reduce
the computational complexity. The INSGA-II is adopted to obtain the Pareto optimal solutions by
Citation: Wang, Y.; Zhou, J.; Sun, Y.;
using the CS insertion operation to select CS locations and integrating the elite retention mechanism
Wang, X.; Zhe, J.; Wang, H. Electric to ensure a stable and excellent performance. The superiority of the hybrid algorithm is proven by
Vehicle Charging Station comparison with the other three algorithms (i.e., multi-objective genetic algorithm, multi-objective
Location-Routing Problem with Time particle swarm optimization, and multi-objective ant colony optimization). An empirical study of the
Windows and Resource Sharing. EVCS-LRPTWRS in Chongqing City, China is conducted. Then, four types of service period divisions
Sustainability 2022, 14, 11681. and three scenarios of resource sharing modes are further analyzed and discussed. The empirical
https://doi.org/10.3390/ results demonstrate the validity and practicability of the proposed solution method in realizing a
su141811681
sustainable operation in EV distribution networks.
Academic Editor: Mouloud Denai
Keywords: charging station location-routing problem; multi-depot multi-period electric vehicle
Received: 16 August 2022
routing optimization; resource sharing; bi-objective nonlinear programming model; hybrid algorithm
Accepted: 14 September 2022
Published: 17 September 2022

Publisher’s Note: MDPI stays neutral


with regard to jurisdictional claims in 1. Introduction
published maps and institutional affil-
Company decision-makers have focused on the concepts of sustainable operations and
iations.
green transportation due to the increasing concern about global warming and environmen-
tal pollution [1,2]. Electric vehicles (EVs) are low-carbon transportation devices that can
reduce harmful emissions and dependence on fossil fuels. These EVs are also a promising
Copyright: © 2022 by the authors.
tool for developing sustainable logistics distribution networks [2,3]. However, two main
Licensee MDPI, Basel, Switzerland. challenges exist in the implementation of EV distribution. On the one hand, the lack of
This article is an open access article charging infrastructure hinders the spread of the EV distribution [4]. On the other hand,
distributed under the terms and the restriction of EVs caused by a limited driving range and necessary charging time, and
conditions of the Creative Commons the growing demand for goods distribution, promote the occurrence of an unreasonable EV
Attribution (CC BY) license (https:// scheduling [4–6]. Therefore, optimizing the EV charging station (CS) location routing prob-
creativecommons.org/licenses/by/ lem (EVCS-LRP) by integrating the transportation resource sharing strategy and service
4.0/). periods division is crucial to improving the efficiency of the EV distribution networks.

Sustainability 2022, 14, 11681. https://doi.org/10.3390/su141811681 https://www.mdpi.com/journal/sustainability


Sustainability 2022, 14, 11681 2 of 31

EV routing planning depends on the availability of CSs, and the selection of CS


locations will be determined according to the delivery routes. This notion means that
integrating EV scheduling into CS locations is essential [7,8]. The optimization of the
locations of CSs and EV routes reduces the construction cost of EV distribution networks [9].
Furthermore, a transportation resource sharing strategy, as an effective method to optimize
logistics operations in the urban freight industry, can reduce the number of required
EVs, thus decreasing the environmental impact and avoiding traffic congestion [10,11].
Meanwhile, the division of multiple service periods in the whole planning horizon to
further facilitate resource sharing among multiple depots, contributes to the rationalization
of resource allocation and improvement of the efficiency and sustainability of the EV
distribution networks [12,13].
The EV CS location-routing problem with time windows and resource sharing (EVCS-
LRPTWRS) is presented and solved in this study. The CS location decisions and EV routing
schemes are optimized among multiple depots within various service periods, and the
transportation resource sharing strategy is applied to rationalize the resource configuration.
A bi-objective nonlinear mathematical programming model with multiple service periods is
formulated to minimize the total operating cost (TOC) and the number of required EVs. A
hybrid algorithm composed of the Gaussian mixture clustering algorithm (GMCA) and an
improved nondominated sorting genetic algorithm-II (INSGA-II) is proposed to address the
EVCS-LRPTWRS. To simplify the logistics network optimization, customers are assigned to
appropriate depots and divided into suitable service periods according to the GMCA. Then,
the Pareto optimal solutions, including the optimal delivery routes and the CS location
decisions are obtained by integrating the CS insertion operation into the NSGA-II. The
application of the EVCS-LRPTWRS is verified by an empirical study. The results show that
the proposed nonlinear programming model and hybrid algorithm can effectively optimize
the CS location-routing problem and improve the EV delivery efficiency.
The contributions of this study are as follows: (1) A multi-depot multi-period EV
distribution network is presented according to the periodic characteristics of the de-
pots and customers, to determine the CS locations and optimize the EV delivery routes.
(2) A transportation resource sharing strategy is integrated into the EVCS-LRPTWRS to
improve the transportation efficiency and optimize resource configuration among multiple
depots within various service periods. (3) A bi-objective nonlinear programming model
integrating multiple service periods, a transportation resource sharing strategy, and a
centralized transportation is constructed to minimize the operating cost and number of
EVs. (4) A hybrid algorithm consisting of the GMCA and INSGA-II is developed to solve
the bi-objective nonlinear programming model and obtain the Pareto optimal solutions
including the CS location decisions and EV routes.
The remainder of this study is organized as follows. The related literature is described
in Section 2. The problem of the EVCS-LRPTWRS is presented in Section 3. The related
definitions and model formulation are elaborated in Section 4. The hybrid algorithm is
developed in Section 5. The computational experiments are illustrated in Section 6. The
conclusion and future research directions of this study are summarized in Section 7.

2. Literature Review
Recently, an increasing number of academic experts and enterprisers have started to
pay attention to EVCS-LRP research due to the growing environmental awareness [14,15].
Moreover, EVCS-LRP-related research has grown in complexity. For example, common
constraints such as multi-depot, time windows, and multi-period, are added to the stan-
dard EV location routing problem (EVLRP) to form various variants [16,17]. Meanwhile,
transportation resource sharing is studied by an increasing number of researchers as a
promising strategy to reduce the economic and environmental impacts of traffic conges-
tion [10,11]. Researchers have developed a variety of related methods to address the
complex problem [17,18].
Sustainability 2022, 14, 11681 3 of 31

2.1. Electric Vehicle Charging Station Location-Routing Problem


EVCS-LRP refers to simultaneously solving the CS location selection and EV routing
optimization problems [7,19,20]. Yang and Sun [21] formulated an integer programming
model to simultaneously decide the locations of battery swap stations (BSSs) and the route
plan of EVs. Yıldız et al. [22] developed an enhanced deviation flow refueling location
model to handle the refueling location station problem with routing by setting a tolerable
shortest routing deviation. Schiffer and Walther [7] discussed the impact of different
objective functions, such as the distance, the number of EVs and the required CSs, and
the cost minimization, on the EV routing plan and CS location decisions. He et al. [23]
proposed a bi-level programming model to determine the CS locations by maximizing the
routing flow and balancing the routing plan under the constraint of the driving distance.
Zhang et al. [24] introduced two resource strategies based on battery and loading capacities
to determine the optimal number and locations of BSSs and the routes for the BSS location-
routing problem. Guo et al. [25] developed a mathematical planning model based on user
satisfaction and the detour limitations to solve the selection of service station locations
and route planning for EVs. Çalık et al. [26] constructed a mixed integer programming
model to minimize the total cost, including the routing costs and operating costs of CSs and
EVs, in determining the location of CSs and the routes for EVs. Yang et al. [1] presented
an optimization model based on a space time state energy representation network in the
EVCS-LRP to reduce the operating cost and improve efficiency.

2.2. Electric Vehicle Location-Routing Problem and its Variants


The EVLRP and its variants, such as multi-depot, time windows, and multi-period,
have been extensively studied in recent years [16,17]. Wang et al. [27] studied a multi-depot
EV scheduling problem to minimize the total costs, including the fixed costs, travel costs
and charging costs (CC) of EVs. Sadati and Çataya [28] defined a mixed integer linear
programming model to address a multi-depot green vehicle routing problem (MDGVRP).
The location-routing problem with time windows has been studied due to the significant
influence of time windows on evaluating the efficiency of EV distribution networks [29,30].
Ghobadi et al. [2] translated the time constraints into cost in terms of economy and customer
satisfaction to penalize the time window violations in the multi-depot EV routing problem
(EVRP). Lam et al. [29] established an electric vehicle problem with time windows and
considered the limitations of the CS capacities and the piecewise-linearity of charging times.
Furthermore, a multi-period location routing problem was studied to satisfy the different
time requirements of customers [12,13]. Li et al. [31] established a multi-period, multi-route
CS location model to obtain the cost-effective location scheme from the spatial and temporal
dimensions. Neves-Moreira et al. [32] proposed a novel mathematical formulation to deal
with the multi-period vehicle routing problem with refueling decisions. Lin et al. [33]
explored the impact of different service periods on the optimization results for the multi-
period EV routing optimization problem.

2.3. Transportation Resource Sharing Strategy


Transportation resource sharing, which is an effective strategy to save logistics re-
sources and costs where improving the efficiency of logistics networks, is conducive to
the sustainable development of logistics [13,34]. Brandstätter et al. [34] combined the
respective strengths of car sharing and EVs to overcome the challenges of traffic conges-
tion and environmental pollution. Friedrich and Noekel [35] discussed the advantages of
vehicle sharing systems in public transport networks and provided an assignment model
for solving a vehicle sharing problem. Wang et al. [13] incorporated the transportation
resource sharing strategy into the MDGVRP to eliminate long empty trips and improve
the flexibility of the network and the efficiency of resource management. Almouhanna
et al. [36] allowed all EVs to be shared among the depots in the electric location routing
problem to save resources. Jin et al. [37] proposed a sharing mode of EVs and studied the
various pricing strategies to improve the utilization rate of shared vehicles. Wang et al. [12]
Sustainability 2022, 14, 11681 4 of 31

demonstrated that transportation resource sharing among multiple service periods can
maximize resource utilization.

2.4. Relevant Solution Methods for Electric Vehicle Location-Routing Problem


Researchers consider the EVLRP to be a NP-hard problem that is primarily addressed
via heuristic algorithms [18,38], machine learning [39,40], and evaluation methods [41,42].
Chen et al. [38] developed a heuristic branch-and-price algorithm with an adaptive selec-
tion scheme to simultaneously determine the locations of BSSs and the routing plan. Barco
et al. [43] presented a differential evolution algorithm to coordinate the EV routing and
recharge scheduling for solving the EVRP. Kyriakakis et al. [44] proposed an ant colony opti-
mization algorithm as a solution method for the EVRP to minimize the energy consumption
of EVs. Zhang et al. [45] incorporated a Gaussian variation operator, a differential evolution
operator, and a congestion factor into the whale optimization algorithm to solve the CS
location problem of EVs. Yu et al. [46] developed an adaptive larger neighborhood search
(ALNS) algorithm to improve performance by specific removal and insertion operations of
CSs in solving the variant of the EVRP with a mixed fleet. Erdem [47] designed an adaptive
variable neighborhood search (AVNS) to optimize the routing with heterogeneous EVs
for a sustainable urban recycling waste collection. Table 1 lists the abbreviations of the
relevant literature for the EVCS-LRPTWRS and Table 2 summarizes the relevant studies for
the EVCS-LRPTWRS.

Table 1. Abbreviations of the relevant literature for the EVCS-LRPTWRS.

Abbreviation Type of Problem Abbreviation Solution Method


Electric vehicle location-routing problem
EVLRPTW VNS Variable neighborhood search
with time windows
Multi-depot electric vehicle location routing problem Multi-objective particle
MDEVLRPTW MOPSO
with time windows swarm optimization
Electric vehicle location-routing problem with a
EVLRP-BSS LNS Large neighborhood search
battery swap station
E2EVRP Two-level electric vehicle routing problem MOHA Multi-objective heuristic algorithm
GVRP-PR Green vehicle routing problem with partial charging CG Column generation
Electric vehicle routing problem with a
EVRP-BSS BP Biased-randomized
battery swap station
EVRPTW Electric vehicle routing problem with time windows PSO Particle swarm optimization

In Table 2, a comparison of 20 previous papers with the proposed study was pre-
sented. The comparison content is divided into the following five aspects: (1) type of
problems; (2) variants of the multi-depot, multi-echelon, time windows, and multi-period;
(3) type of electric fleet: homogeneous or heterogeneous; (4) resource sharing strategy; and
(5) solution method. According to the comparison results, the aspects considered in this
study are comprehensive.
The limitations of the aforementioned literature for the EVCS-LRPTWRS optimization
are as follows: (1) CS location decisions and EV routes in the multi-depot multi-period
logistics network design have insufficiently been discussed. (2) The transportation resource
sharing strategy is rarely considered in the optimization of the EVCS-LRP. (3) An effective
mathematical model for a resource sharing strategy and centralized transportation in
the optimization of the EVCS-LRPTWRS is lacking. (4) Existing heuristic algorithms
have limited applicability in solving the EVCS-LRPTWRS within multiple service periods
considering a shared transportation resource.
Sustainability 2022, 14, 11681 5 of 31

Table 2. Relevant literature overview for the EVCS-LRPTWRS.

Variants Electric Fleet

Type of Problems

Resource Sharing

Solution Method
Time Windows

Heterogeneous
Multi-Echelon

Homogeneous
Multi-Period
Multi-Depot
References

√ √
Yang et al. [1] EVLRP √ √ Lagrange relaxation
Schiffer and Walther [7] EVLRPTW √ √ √ √ √ VNS
Wang et al. [12] MDGVRP √ √ √ MOPSO
Paz et al. [20] MDEVLRPTW √ CPLEX
Yang and Sun [21] EVLRP-BSS √ ALNS
Zhang et al. [24] EVLRP-BSS √ √ Hybrid VNS-PSO
Guo et al. [25] EVLRP √ √ K-shortest ALNS
Çalık et al. [26] EVLRP √ Integer programming
Brandstätter et al. [34] CS location problem √ √ √ Heuristic algorithm
Almouhanna et al. [36] EVLRP √ √ BR-VNS
Breunig et al. [48] E2EVRP √ LNS
Chakraborty et al. [49] EVRP √ MOHA
Felipe et al. [50] GVRP-PR √ √ VNS
Froger et al. [51] EVRP √ √ Labeling algorithm
Jie et al. [52] EVRP-BSS √ √ CG-ALNS
Hof et al. [53] EVLRP-BSS √ √ √ AVNS
Ma et al. [54] EVRP √ √ ALNS
Pelletier et al. [55] EVRP √ LNS
Zhu et al. [56] CS location problem √ √ √ Genetic algorithm
Keskin and Çatay [57] EVRPTW √ √ √ √ √ √ ALNS
This work EVCS-LRPTW INSGA-II

3. Problem Statement
The optimization of the EVCS-LRPTWRS is advantageous in designing an efficient lo-
gistics network and reducing the environmental impact of the logistics operations [13,36,58].
The EVCS-LRPTWRS aims to simultaneously deal with the location selection of CSs and
the routing optimization in the multi-depot multi-period EV distribution network. The EV
distribution network consists of multiple depots, several candidate CSs, and a number of
customers. Multiple depots are connected by electric trucks (ETs), and customers are served
by EVs from depots. The number and location of CSs should be determined and selected
from the candidate CSs to satisfy the charging needs of EVs during the delivery process.
In addition, a working period can be divided into several service periods based on the
different time windows of depots and customers. Specifically, each depot has several oper-
ating time windows, and every customer has a service time window during a work period.
Figure 1 illustrates the comparison of the EVCS-LRPTWRS before and after optimization.
Figure 1a shows the initial EV distribution network before optimization. In this study,
EVs need a significant amount of power on the way to serve customers due to the in-
appropriate use of the CSs, resulting in a large portion of electricity consumption (EC)
and waste of transportation resources. Moreover, the irrational resource configuration
inevitably leads to the occurrence of long-distance and crisscross transportation, thus
making considerable violations of customers’ time windows. The independence of the
delivery routes between two service periods of depots causes a large number of required
EVs. Figure 1b shows the optimized CS location-routing network. A more orderly EV
distribution network without long-haul and overlapping delivery routes is organized by
the optimized location selection of CSs. The sets of CSs selected to provide charging ser-
vices for EVs in the two service periods are {CS2, CS4} and {CS4, CS6}, and the location
selections of CSs are independent between the two service periods. In comparison with
the initial EV distribution network, an ET can transport goods to multiple depots while
Sustainability 2022,
Sustainability 2022, 14,
14, 11681
x FOR PEER REVIEW 66of
of 32
31

a numberthe
meeting oftime
customers.
windows Multiple depotsand
of all depots, are the
connected
numberby ofelectric trucks
customers (ETs),
served and cus-
beyond the
tomers are
required served
time by EVshas
windows from depots.
been The number
significantly and location
reduced. of CSs
In addition, should
each be deter-
EV can cover
mined and
several selected
routes from
to cover thethe candidate
share CSs to satisfy
of transportation the charging
resources amongneeds of EVs service
the various during
the delivery
periods. process. In
For example, routes D3→aC17
addition, →C16→
working CS4→can
period →D3
C15be divided
and D3 →C18
into →C19
several →D3
service
are visited
periods by EV3
based within
on the the first
different timeservice
windowsperiod, and EV3
of depots andcontinues
customers.to Specifically,
implement route
each
D2 → C12 → C13 → CS6 → C11 → D2 within the second service period. Therefore,
depot has several operating time windows, and every customer has a service time win- the opti-
mization
dow during of the CS location-routing
a work period. Figurenetwork and transportation
1 illustrates the comparison resource
of thesharing strategy is
EVCS-LRPTWRS
effective
before andin improving the utilization of transportation resources and achieving EC savings.
after optimization.

(+2) [2,6] [2,6]


2 [2,6] (+2) [2,6] (+2)
3 1 [2,6] 1 3 1 [2,6]
[1,3] 2 1 2 1
EV4 4 CS3 [1,3] 4 CS3
CS1 1 EV1 3 EV3
1 [12,21] CS2 [4,9] 2 CS1 1 [12,21] CS2 [4,9]
1 1 5 1 [13,22]
1
[12,15] 1 1 [0,10] 1 [0,9][13,15] [12,15] 1 [0,10] EV1 1 5 1 [0,9] 1
[4,9] 1 [13,15]
8 D1 EV5 D2 1
12 8 D1 3 D2 1 12
1 EV2 1 EV3 3 6 3 1 EV1 1 6 [13,22]
[15,19] EV6 1 1 1 [4,9] 1
2 7 [4,9] 11 2 7 11 1
9 10 9 [16,19] 10 EV2
[12,15] [16,19] 2 1 3 13 [12,15] 1 1 [4,9] [15,19] 1 (+2) 1 13
(+2) [5,8]
2 15EV8 [18,21] 3 [13,15] (+2) 1 15 3 [18,21] 3 [13,15]
(+2)
1 [5,8] CS4 2 CS6
CS4 3 14 EV7
14
1 CS6 1
[2,5] 1 [2,5] 1 EV2 EV3 1
2 EV10 1
16 CS5 3 16 CS5 [0,11] [12,20] CS7
EV9 1 D3 1 21 CS7 D3 21
(+2) 1 [12,16] 1 1 1 1
17 1 [0,11] 1 20 [1,3] 17 1 1 [12,16]
[12,20] EV2 20
1 [1,3] 18 CS8 18
1 EV11 19 [12,15] 19 [12,15]
CS8 [6,9] [6,9] [6,9] 1 [6,9]

(a) Initial EV distribution network (b) Optimized CS location-routing network


Depot Customer Battery level of EVs (+2) Charging time Unselected CS location
Selected CS location Centralized transportation EV distribution routes of period 1
EV distribution routes of period 2 Sharing routes for delivery
[•,•] Time windows of period 1 [•,•] Time windows of period 2 Violation of time window

Figure 1. Illustration of the EVCS-LRPTW optimization.

We assume
Figure that the
1a shows theECinitial
of EVsEVis distribution
only related to unit time,
network andoptimization.
before 8 kWh per unit In time
this
for EC, USD 3 per kWh for electricity prices, USD 15 per time unit for
study, EVs need a significant amount of power on the way to serve customers due to penalty cost (PCs)
the
(earliness and use
inappropriate delay penalties),
of the USD 5 in
CSs, resulting per time unit
a large for of
portion CC, and USDconsumption
electricity 25 per time (EC)
unit
for
andcentralized transportation
waste of transportation cost (CTC).
resources. Table 3the
Moreover, compares tenresource
irrational indicators related to the
configuration in-
optimization
evitably leadsoftothe
theEVCS-LRPTWRS, including the
occurrence of long-distance andCTC, electrictransportation,
crisscross delivery cost (EDC), EC,
thus mak-
PC, CC, TOC, number
ing considerable of EVs,
violations of number
customers’of ETs,
timeand numberThe
windows. of selected CSs. of the deliv-
independence
ery routes between two service periods of depots causes a large number of required EVs.
Table 3. Comparison in the initial and optimized network of the EVCS-LRPTWRS.
Figure 1b shows the optimized CS location-routing network. A more orderly EV distribu-
Service CTC tion network
EDC EC without CC long-haul
Waitingand
andoverlapping
PC delivery
TOC routes
Numberis organized
Number bySelected
the op-
Case timized location selection of CSs. The sets of CSs selected to provide charging services
Period (USD) (USD) (kWh) (USD) Delay Times (USD) (USD) of EVs of ETs CSs for
1st 0
EVs
672
in the224
two service
100
periods 18are {CS2, CS4}270
and {CS4,
1042
CS6}, and
6
the location
0
selections
4
Initial 2nd 0 of
672CSs are224
independent
70 between 4 the two service
60 periods.
802 In comparison
5 0with the initial
2
Total 0 EV
1344distribution
448 network,
170 an ET22can transport 330goods to multiple11
1844 depots while0 meeting 6 the
1st 225 time
456 windows152 of all 70 depots, and1 the number 15 of customers
766 served
3 beyond 1 the required
2
Optimized 2nd 0 time
384 windows
128 has been
60 significantly
0 reduced.0 In addition,
444 each
3 EV can0 cover several
2
Total 225 routes
840 to cover
280 the130share of transportation
1 resources
15 among the3 various service
1210 1 periods.
3
For example, routes D3→C17→C16→CS4→C15→D3 and D3→C18→C19→D3 are visited
by EV3 within the first service period, and EV3 continues to implement route
In Table 3, the total cost and number of EVs were reduced by designing a CS location-
D2→C12→C13→CS6→C11→D2 within the second service period. Therefore, the optimi-
routing network. The TOC decreases from USD 1844 to USD 1210 for the two service
zation of the CS location-routing network and transportation resource sharing strategy is
periods through the optimization of the EVCS-LRPTWRS, achieving a total saving of USD
effective in improving the utilization of transportation resources and achieving EC sav-
634. The number of EVs decreases from 11 to 3, with a drop of 8. The number of selected
ings.
CSs decreases from 6 to 3, with a reduction of 3. Furthermore, the EC decreases from
We assume that the EC of EVs is only related to unit time, and 8 kWh per unit time
448 kWh to 280 kWh, with an electricity-saving of 168 kWh. Therefore, the EVCS-LRPTWRS
for EC, USD 3 per kWh for electricity prices, USD 15 per time unit for penalty cost (PCs)
optimization can effectively reduce the operating cost and improve the utilization of the
(earliness and delay
transportation resourcepenalties), USD
for the EV 5 per timenetwork.
distribution unit for CC, and USD 25 per time unit for
Sustainability 2022, 14, 11681 7 of 31

4. Related Definitions and Model Formulation


4.1. Assumptions and Definitions
In this section, we develop a bi-objective nonlinear programming model to solve the
charging infrastructure location problem integrated with the planning of the EV distribution
routing problem. To clarify the EVCS-LRPTWRS model, the relevant notations and variables
used in the model are defined in Table 4. Additionally, several rational assumptions of the
nonlinear programming model based on the fundamental assumptions in the traditional
EVRP and the multi-depot location-routing problem are stated as follows:

Table 4. Notations and descriptions used in the EVCS-LRPTW model.

Sets Definitions
D Set of depots, D = {1, 2, 3, . . . , d}, d∈D
C Set of customers, C = {1, 2, 3, . . . , c}, c∈C
R Set of CSs, R = {1, 2, 3, . . . , r}, r∈R
T Set of electric trucks for centralized transportation, T = {1, 2, 3, . . . , t}, t∈T
V Set of EVs for delivery routes, V = {1, 2, 3, . . . , v}, v∈V
W Set of service periods within one working period, W = {1, 2, 3, . . . ,w}, w∈W
Mtw The sequence of routes executed by ET t within the wth service period, Mtw = {1, 2, 3, . . . , m}, m ∈ Mtw , t∈T, w∈W
Nvw The sequence of routes executed by EV v within the wth service period, Nvw = {1, 2, 3, . . . , n}, n ∈ Nvw , v∈V, w∈W
Uvw Set of EVs used to serve customers within the wth service period, v∈V, w∈W
NUvnw Set of customers served by EV v in the nth route within the wth service period, n ∈ NUvnw , v∈V, w∈W

Parameters Description
fe Electricity price, (unit: USD/kWh)
Fcc CC of electricity per unit time, (unit: USD/h)
FRr Unit operating cost of CS
δv Charging rate of EV v, v∈V, (unit: kWh/h)
µd Variable cost coefficient of depot d, d∈D
MNv Annual maintenance cost of EV v, v∈V
K Number of working periods in one year
w EC of EV v from node d to c in the nth route within the wth service period, d, c ∈ R ∪ D ∪ C, d 6= c, v ∈ V, n ∈ Nvw ,
ECVdcvn
w∈W, (unit: kWh)
w EC of ET tfrom depot d to d0 in the mth route within the wth service period, d, d0 ∈ D, d 6= d0 , v ∈ V, m ∈ Mtw ,
ECTdd 0tm w∈W,(unit: kWh)
αv Arc specific coefficient of EV v, v∈V
λt Arc specific coefficient of ET t, t∈T
γt Vehicle specific coefficient of ET t, t∈T
βv Vehicle specific coefficient of EV v, v∈V
SVdn w Driving speed of EV vin the nth route within the wth service period, v ∈ V, n ∈ Nvw , w∈W, (unit: m/s)
STtm w Driving speed of ET tin the mth route within the wth service period, t ∈ T, m ∈ Mtw , w∈W, (unit: m/s)
WEVvn w Weight of EV v in the nth route within the wth service period, v ∈ V, n ∈ Nvw , w∈W, (unit: kg)
WETtm w Weight of ET t in the mth route within the wth service period, t ∈ T, m ∈ Mtw , w∈W, (unit: kg)
ldc Distance from node d to c, d, c∈R∪D∪C, d 6= c
Ldd’ Distance from depot d to d’, d, d’∈D, d 6= d’
QTt Load capacity of ET t, t∈T
QVv Load capacity of EV v, v∈V
∂wc Demand quantity of customer c within the wth service period, c∈C, w∈W
qwdd0 Transport quantity from depot d to d’ within the wth service period, d, d’∈D, d 6= d’, w∈W
qd Demand quantity of depot d, d∈D
Pe PC per unit of time of earliness
Pl PC per unit of time of delay
[edw , ldw ] Operation time window for depot d within the wth service period, d∈D, w∈W
[ gcw , hwc ] Service time window for customer c within the wth service period, c∈C, w∈W
w Departure time of EV v from depot d in the nth route within the wth service period, d ∈ D, v ∈ V, n ∈ Nvw , v∈V,
LTdvn
w∈W
w
DTdtm Departure time of ET t from depot d in the mth route within the wth service period, d ∈ D, t ∈ T, m ∈ Mtw , w∈W
w
LTcvn Departure time of EV v from customer c in the nth route within the wth service period, c ∈ C, v ∈ V, n ∈ Nvw ,w∈W
Sustainability 2022, 14, 11681 8 of 31

Table 4. Cont.

Sets Definitions
w
LTrvn Departure time of EV v from CS r in the nth route within the wth service period, r ∈ R, v ∈ V, n ∈ Nvw , w∈W
w
ATdvn Arrival time of EV v at depot d in the nth route within the wth service period, d ∈ D, v ∈ V, n ∈ Nvw , w∈W
w
GTdtm Arrival time of ET t at depot d in the nth route within the wth service period, d ∈ D, t ∈ T, m ∈ Mtw , w∈W
w
ATcvn Arrival time of EV v at customer c in the nth route within the wth service period, c∈C, v∈V, n∈ Nvw , w∈W
w
ATrvn Arrival time of EV v at CS r in the nth route within the wth service period, r ∈ R, v ∈ V, n ∈ Nvw , w∈W
w Waiting time of EV v at customer c in the nth route within the wth service period, and the waiting time denotes the time
WTcvn
required to wait for the service to start, c ∈ C, v ∈ V, n ∈ Nvw , w∈W
w
CTrvn Charging time of EV v at CS r in the nth route within the wth service period, r ∈ R, v ∈ V, n ∈ Nvw , w∈W
w Travel time of EV v from CS r to customer c in the nth route within the wth service period, r ∈ R, c ∈ C, v ∈ V,
JTrcvn
n ∈ Nvw ,w∈W
w Travel time of EV v from customer d to CS r in the nth route within the wth service period, r∈R, d∈C, v∈V,n∈ Nvw ,
JTdrvn
w∈W
w Travel time of EV v from customer d to c in the nth route within the wth service period, d, c∈C, d 6= c, v∈V, n∈ Nvw ,
JTdcvn
w∈W
w
LErvn Amount of electricity remaining when EV v leaves depot or CS r, r∈R∪D, v∈V, n∈ Nvw , w∈W, (unit: kWh)
w
LEdvn Amount of electricity remaining when EV v leaves customer d, c ∈ C, v ∈ V, n ∈ Nvw , w∈W, (unit: kWh)
w
LEcvn Amount of electricity remaining when EV v leaves customer c, c∈C, v∈V, n∈ Nvw , w∈W, (unit: kWh)
w
AEcvn Amount of electricity remaining when EV v arrives at customer c, c ∈ C, v ∈ V, n ∈ Nvw , w∈W, (unit: kWh)
w
RErvn Amount of electricity remaining of EV v at node r, r ∈ R ∪ D ∪ C, v ∈ V, n ∈ Nvw , w∈W, (unit: kWh)
w
CEdrvn Amount of EC of EV v from customer d to CS , c ∈ C, r ∈ R, v ∈ V, n ∈ Nvw , w∈W, (unit: kWh)
w
CErcvn Amount of EC of EV v from CS to customer c, c ∈ C, r ∈ R, v ∈ V, n ∈ Nvw , w∈W, (unit: kWh)
w
CEdcvn Amount of EC of EV v from customer dto c , d, c ∈ C, d 6= c, v ∈ V, n ∈ Nvw , w∈W, (unit: kWh)
Ev Battery capacity of EV v, v∈V, (unit: kWh)
MM Extremely large number
| Nvw | Number of routes assigned to EV v within the wth service period
| NUvn w| Number of customers served by EV v in the nth route within the wth service period
w
|Uv | Number of EVs used to serve customers within the wth service period
Decision
Description
variable
If EV v travels from node d to node c in the nth route within the wth service period xdcvn w = 1, otherwise, x w = 0,
w
xdcvn dcvn
w
d, c ∈ R ∪ D ∪ C, d 6= c, v ∈ V, n ∈ Nv , w∈W
w If ET t travels from depot d to d0 in the mth route within the wth service period πdd w w
0tm = 1, otherwise, πdd0tm = 0,
πdd 0tm d, d0 ∈ D, d 6= d0 , t ∈ T, m ∈ Mtw , w∈W
w = 1; otherwise, x w = 0, d, c∈R∪D∪C, d 6 = c,
If EV v travels from node d to node c within the wth service period, xdcv
w
xdcv dcv
v∈V, w∈W
ϕw
v If EV v is selected to serve customers within the wth service period, v∈V, w∈W
Odcd0 If customer c is reassigned from depot d to d0 , Odcd0 = 1; otherwise, Odcd0 = 0, d, d, d’∈D, d 6=d’, c∈C, v∈V
yrv If EV v charged at CS or depot r yrv = 1; otherwise, yrv = 0, r∈R∪D, v∈V

Assumption 1. The demands, locations, and time windows of customers are known, the service
time is ignored for all customers and they are visited exactly once within one service period.
Assumption 2. Each EV departs from one depot and finally returns to the same depot.
Assumption 3. EVs depart from depots or CSs with a full battery.
Assumption 4. EVs can be charged at any electricity level, and their charging and driving speeds
are constant.

4.2. Model Formulation


A bi-objective nonlinear programming model, which includes the minimization of
the operating cost and the number of EVs, is developed to solve the EVCS-LRPTWRS.
The operating cost objective function is established in Equation (1), which contains three
components, namely TC1, TC2, and TC3. The number of the EV objective function is
presented as Equation (2).
Sustainability 2022, 14, 11681 9 of 31

Min Z1 = TC1 + TC2 + TC3 (1)


Min Z2 = max{ ∑ ϕw
v · (min{ ∑ ∑ ∑ w
xdcvn , 1})} (2)
w ∈W v ∈ V
d∈ D c∈C n∈ Nvw

In Equation (3), TC1 indicates the EDC within a working period. f e · ∑ ∑


d,c∈ D ∪C ∪ R,d6=cv∈V
w · xw
∑ ∑ ECVdcvn dcvn is the EC cost generated by EVs traveling between depots and cus-
w∈W n∈ Nvw
w can be calculated by Equation (4). F ·
tomers, and the EC of EVs. ECVdcvn cc ∑ ∑ ∑ ∑ yrv ·
r ∈ R v∈V w∈W n∈ Nvw
w )
( Ev − RErvn w )
( Ev − RErvn
δv is the CC of EVs and δv is the charging time.
w )
( Ev − RErvn
TC1 = f e · ∑ ∑ ∑ ∑ w ECVdcvn
w w
· xdcvn + Fcc · ∑ ∑ ∑ ∑ yrv ·
w δv
(3)
d,c∈ D ∪C ∪ R,d6=c v∈V w∈W n∈ Nv r ∈ R v∈V w∈W n∈ Nv

w w w2
ECVdcvn = (αv · WEVvn + β v · SVvn ) · ldc (4)
In Equation (5), TC2 denotes the PC of earliness or delay for EVs.

TC2 = ∑ ∑ Pe · (max{ gcw − ∑ ∑ w ATcvn


w
, 0}) + ∑ ∑ Pl · (max{ ∑ ∑ w ATcvn
w
− hw
c , 0}) (5)
c ∈ C w ∈W v∈V n∈ Nv c ∈ C w ∈W v∈V n∈ Nv

In Equation (6), TC3 represents the operating cost of facilities and the maintenance
cost of EVs in one working period. ∑ Fd and ∑ µd · qd are the fixed and variable costs of
d∈ D d∈ D
w w
depots, respectively. f e · ∑ ∑ ∑ ∑ ECTdd 0tm · πdd0tm is the CTC among depots
d,d0∈ D,d6=d0 t∈ T w∈W m∈ Mtw
w MNv
and the EC of ETs. ECTdd0tm can be calculated by Equation (7). ∑ K · max{|Uvw |} is the
v ∈V w ∈W
maintenance cost of EVs, and FRr · ∑ yrv is the operating cost of the selected CSs.
r∈R

TC3 = ∑ Fd + ∑ µd · qd + f e · w w
∑ ∑ ∑ ∑ ECTdd 0tm · πdd0tm
d∈ D d∈ D d,d0∈ D,d6=d0 t∈ T w∈W m∈ Mtw
MNv (6)
+∑ K · max {|Uvw |} + FRr · ∑ yrv
v ∈V w ∈W r∈R

w w w
ECTdd 0tm = ( λt · WETtm + γt · STtm ) · Ldd0 (7)
Subject to:
∑ ∑ ∑ w xdcvn
w
= 1, c ∈ C, w ∈ W (8)
d∈ D ∪C ∪ R v∈V n∈ Nv

∑ w
xdcvn − ∑ w
xcdvn = 0, c ∈ C, v ∈ V, n ∈ Nvw , w ∈ W (9)
d∈ D ∪C ∪ R d∈ D ∪C ∪ R

∑ v · (min{ ∑
ϕw ∑ ∑ w xdcvn
w
, 1}) = |Uvw |, w ∈ W (10)
v ∈V d∈ D c∈C n∈ Nv

∑ ∑ xdcv
w
= | Nvw |, v ∈ V, w ∈ W (11)
d∈ D c∈C

∑ ∑ xdcvn
w
≥ ∑ ∑ xdcv
w w w
(n+1) , v ∈ V, n ∈ Nv , n 6 = | Nv |, w ∈ W (12)
d∈ D c∈C d∈ D c∈C

∑ ∑ qwc · xdcvn
w
≤ QVv , c ∈ C, v ∈ V, n ∈ Nvw , w ∈ W (13)
d∈ D ∪C ∪ R c∈C

∑ qw w w
dd0 · πdd0tm ≤ QTt , t ∈ T, m ∈ Mt , w ∈ W (14)
d,d0∈ D,d6=d0
Sustainability 2022, 14, 11681 10 of 31

qw
dd0 = ∑ Odcd0 · ∂wc , d, d0 ∈ D, d 6= d0, w ∈ W (15)
c∈C

∑ w
w
xdcvn w
≤ | NUvn | − 1, v ∈ V, n ∈ Nvw , w ∈ W (16)
d,c∈ NUvn

edw · ∑ xdcvn
w w
≤ LTdvn ≤ ldw · ∑ xdcvn
w
, d ∈ D, v ∈ V, n ∈ Nvw , w ∈ W (17)
c∈C c∈C

edw · ∑ xdcvn
w w
≤ ATdvn ≤ ldw · ∑ xdcvn
w
, d ∈ D, v ∈ V, n ∈ Nvw , w ∈ W (18)
c∈C c∈C

edw · ∑ w
πdd w w
0tm ≤ DTdtm ≤ ld · ∑ w
πdd w
0tm , d ∈ D, t ∈ T, m ∈ Mt , w ∈ W (19)
d0∈ D d0∈ D

edw · ∑ w
πdd w w
0tm ≤ GTdtm ≤ ld · ∑ w
πdd w
0tm , d ∈ D, t ∈ T, m ∈ Mt , w ∈ W (20)
d0∈ D d0∈ D
w ≤ GT w − MM · (2 − π w w
DTdvn dtm dd0tm − xdcvn ), (21)
d, d0 ∈ D, d 6= d0, c ∈ C, v ∈ V, t ∈ T, n ∈ Nvw , m ∈ Mtw , w ∈ W

w w w w
ATcvn + WTcvn − MM · (1 − xdcvn ) ≤ LTcvn , d ∈ D ∪ C ∪ R, c ∈ C, v ∈ V, n ∈ Nvw , w ∈ W (22)

w w w w
ATcvn + WTcvn + MM · (1 − xdcvn ) ≥ LTcvn , d ∈ D ∪ C ∪ R, c ∈ C, v ∈ V, n ∈ Nvw , w ∈ W (23)
w w w w
ATrvn + CTrvn − MM · (1 − xcrvn ) ≤ LTrvn , r ∈ R, c ∈ C, v ∈ V, n ∈ Nvw , w ∈ W (24)
w w w w
ATrvn + CTrvn + MM · (1 − xcrvn ) ≥ LTrvn , r ∈ R, c ∈ C, v ∈ V, n ∈ Nvw , w ∈ W (25)
w + ( JT w w w w w w w w w
LTcvn drvn · xdrvn + CTrvn · yrv + JTrcvn · xrcvn ) · (1 − xdcvn ) + JTdcvn · xdcvn (26)
− MM · (1 − x(wc+1)rvn − xcw(c+1)vn ) ≤ ATrvn
w , r ∈ R, d, c ∈ C, v ∈ V, n ∈ N w , w ∈ W
v

w + ( JT w w w w w w w w w
LTcvn drvn · xdrvn + CTrvn · yrv + JTrcvn · xrcvn ) · (1 − xdcvn ) + JTdcvn · xdcvn (27)
w w w w
+ MM · (1 − xcrvn − xdcvn ) ≥ ATrvn , r ∈ R, d, c ∈ C, v ∈ V, n ∈ Nv , w ∈ W

∑ ∑ w
xcrvn ≤ MM · yrv , r ∈ R, w ∈ W (28)
c ∈ C v ∈V
w
LErvn = Ev , r ∈ D ∪ R, v ∈ V, n ∈ Nvw , w ∈ W (29)
w w
AEcvn = LEcvn , c ∈ C, v ∈ V, n ∈ Nvw , w ∈ W (30)
w
RErvn ≥ 0, r ∈ D ∪ R ∪ C, v ∈ V, n ∈ Nvw , w ∈ W (31)

AE(wc+1)vn ≤ LEcvn
w + [( E − REw ) · y − CEw
v rvn rv
w w w w w w
crvn · xcrvn − CEr (c+1)vn · xr (c+1)vn ] · (1 − xc(c+1)vn ) − CEc(c+1)vn · xc(c+1)vn ,
(32)
c, c + 1 ∈ C, r ∈ R, v ∈ V, n ∈ Nvw , w ∈ W
w
xdcv = {0, 1}, d, c ∈ D ∪ R ∪ C, d 6= c, v ∈ V, w ∈ W (33)
w
xdcvn = {0, 1}, d, c ∈ D ∪ R ∪ C, d 6= c, v ∈ V, n ∈ Nvw , w ∈ W (34)
w w
πdd 0tm = {0, 1}, d, d 0 ∈ D, d 6 = d 0, t ∈ T, m ∈ Mt , w ∈ W (35)
ϕw
v = {0, 1}, v ∈ V, w ∈ W (36)
Odcd0 = {0, 1}, d, d0 ∈ D, d 6= d0, c ∈ C (37)
yrv = {0, 1}, r ∈ R, v ∈ V (38)
Constraint (8) guarantees that each customer can be served by only a single EV
within one service period. Constraint (9) is the flow conservation during the delivery
process. Constraint (10) defines the number of EVs required to serve customers within
one service period. Constraint (11) defines the number of routes implemented by each
Sustainability 2022, 14, 11681 11 of 31

EV within one service period. Constraint (12) ensures the sequence of routes performed
by each EV. Constraints (13) and (14) limit the total cargo transportation volume to
not exceed the corresponding load capacity of EVs and trucks, respectively. Constraint
(15) expresses that the amount of cargo transported is equal to the total reassigned
customer demands among depots. Constraint (16) is adopted to eliminate subtours.
Constraints (17)–(20) ensure that the departure and return times of EVs and ETs at the
depot are all in the depot’s operating time windows. Constraint (21) guarantees that the
centralized transportation among depots must be completed before serving customers
within one service period. Constraints (22) and (23) express that the departure time
from a customer is equal to the arrival time of EVs plus the waiting time. Constraints
(24) and (25) represent that the departure time from a CS is equal to the arrival time
plus the charging time. Constraints (26) and (27) express that the time relationship
between two nodes, that is, the departure time from a depot plus the travel time and
charging time should be equal to the arrival time at a customer. Constraint (28) stipulates
that that EVs can only be charged at selected CSs. Constraint (29) stipulates EVs must
be charged with a full battery while leaving CSs or depots. Constraint (30) indicates
that the battery level of EVs should remain stable at the departure and arrival of the
same customer. Constraint (31) ensures the non-negativity of the battery level for EVs.
Constraint (32) is the electricity relationship between two customers, that is, the power
leaving customer c plus the power charged at CSs, and minus the power consumed in
travel equals to the power arriving at customer c + 1. Constraints (33)–(38) define the
binary decision variables.

5. Solution Methodology
This section summarizes the basic ideas of the method to solve the EVCS-LRPTWRS
based on the basic framework of the hybrid algorithm. Section 5.1 introduces the GMCA to
reduce the complexity of problem-solving. Section 5.2 describes the INSGA-II to determine
the optimal EV route planning scheme and CS location selections. The proposed heuristic
algorithm is composed of non-dominated ranking, crossover and mutation operations,
elite retention mechanism, and greedy charging strategies. The framework of the hybrid
algorithm is shown in Figure 2.
In Figure 2, a hybrid algorithm that combined the GMCA with INSGA-II is simul-
taneously proposed to solve the vehicle routing optimization and CS location problems.
The GMCA is applied to reduce the complexity of optimizing large logistics networks
by assigning customers to the corresponding depots with a certain probability based on
their coordinates and time windows. The INSGA-II is adopted to design EV distribution
plans and CS selection decisions during the multi-depot multi-period and to find optimal
solutions for the EV distribution networks.

5.1. Customer Clustering


The clustering algorithm based on the specific characteristics of customers (i.e., service
attributes, time windows, and geographic locations) is an efficient method for dealing
with a large-scale logistics network optimization problem [59,60]. The GMCA is applied
in this study to reduce the complexity of the logistics network optimization by dividing a
large network into multiple sub-networks based on the time windows and geographical
locations of customers and depots. The Gaussian mixture model (GMM) is introduced as a
probabilistic model in the GMCA to calculate the probability that each customer belongs to
various depots. The model represents the customer dataset by a weighted combination of
the normal distributions of several depots, and the GMM with k mixture clusters can be
described by Equation (39):
Sustainability 2022, 14, 11681 12 of 31



 ψ(c) = ∑kd=1 wd × φ(c|µd , Σd ) n o

T −1 ( c − µ )
φ(c|µd , Σd ) = √ 1
exp − 1
2 ( c − µ d ) Σ d d (39)
 |Σd |×(2π )i/2
 k w =1
∑ d =1 d

tainability 2022, 14, x FOR PEER REVIEW 13 of
where φ is the Gaussian probability density function; and wd , µd , and Σd are the mixing
factor, mean vector, and covariance matrix of the kth mixture cluster, respectively.

GMCA
Start Calculate the posterior probability
of each mixed cluster
Input: parameters of distribution network (i.e.,
time windows and geographical coordinates of Compute Gaussian mixture and
customers and depots, locations and numbers of posterior distributions
CSs, and information of electric fleets)
Update model parameters
g =1

Determine and divide service periods based on


Check if the new parameters Yes
the time windows of customers and depots
need to be adjusted?
Construct data matrix of customers and depots, and
No
set Gaussian mixed clusters for each service period
Divide customers into corresponding depots
Initialize model parameters of GMCA: mean based on the principle of maximum posterior
vector, mixing factor, and covariance matrix probability and output clustering results

INSGA-II
Greedy insertion strategy for CS selection NSGA-II for vehicle routing plan
No g≤1
Check if the battery level of EV No Yes
is negative for each route? Initialize parameters and generate initial
population Pt based on scanning algorithms
Yes
Find the first customer where the battery r =1
level of EV is negative for the route
Evaluate the fitness function of
Perform CS insertion operation each chromosome and perform
non-dominated sorting
Search for all feasible insertion points
according to battery level of EV before Perform genetic operations: selection, crossover,
the first customer for the route mutation to generate child population Qt

Select a CS from the set of candidate CSs Combine parent population Pt and child
based on the principle of minimum distance population Qt to obtain new population Rt
increment for each feasible insertion point

Calculate and compare the Execute non-dominated sorting and


corresponding delivery cost increment crowding distance comparison to
of each feasible insertion point determine non-dominated ranking

Determine the final insertion point with r = r+1 r≤r_max Yes


the minimum delivery cost increment
and insert selected CS for the route No

g = g+1 g≤g_max Yes


No Check if all routes satisfy the Yes
No
battery limitation
Output Pareto optimal solutions

Figure 2. Flowchart
Figure of of
2. Flowchart thethehybrid algorithm.
hybrid algorithm.

In Figure 2, a hybrid algorithm that combined the GMCA with INSGA-II is simul
neously proposed to solve the vehicle routing optimization and CS location problems. T
GMCA is applied to reduce the complexity of optimizing large logistics networks by
signing customers to the corresponding depots with a certain probability based on th
coordinates and time windows. The INSGA-II is adopted to design EV distribution pla
  (39)
d d  ( c |  ,  ) 
d d (2d )i / 2  2 i / 2 exp
d d (c  d d ) d (c  d )  (3
  d  (2 )  2 
 k 
 d 1 wd  1  k w  1
  d 1 d
Sustainability 2022, 14, 11681 where ϕ is the Gaussian probability density function; and wd, μd, and Σd are the mixing
where ϕ is the Gaussian probability density function; and wd, μd, and Σ13 of 31the mixi
d are
factor, mean vector, and covariance matrix of the kth mixture cluster, respectively.
factor, mean vector, and covariance matrix of the kth mixture cluster, respectively.
The expectation maximization (EM) algorithm is adopted to fit the GMM by first es-
The expectation maximization (EM) algorithm is adopted to fit the GMM by first e
timating the initial parameters and iteratively improving the estimates, thus, estimating
The expectationthe
timating initial parameters
maximization and iteratively
(EM) algorithm improving
is adopted theGMM
to fit the estimates,
by firstthus,
esti-estimati
the values of the parameters to guarantee that the GMM has the maximum likelihood
mating the the values
initial of the parameters
parameters to guarantee
and iteratively improvingthat the GMMthus,
the estimates, has the maximum
estimating the likeliho
[61,62]. The EM algorithm can be divided into the expectation step (E-step) and the maxi-
[61,62].
values of the The EM
parameters toalgorithm
guaranteecan
thatbethe
divided
GMM into the maximum
has the expectationlikelihood
step (E-step) and the ma
[61,62].
mization step (M-step). The E-step is used to calculate and evaluate the probability that
mizationcan
The EM algorithm stepbe(M-step).
divided The
intoE-step is used to step
the expectation calculate andand
(E-step) evaluate the probability th
the maximiza-
the customers belong to the clusters, and the M-step is applied to determine and update
the customers
tion step (M-step). belongisto
The E-step the to
used clusters, andand
calculate the evaluate
M-step isthe applied to determine
probability that theand upda
the parameters. The specific procedure of the GMCA is shown in Algorithm 1.
customers the parameters.
belong The specific
to the clusters, and theprocedure
M-step isofapplied
the GMCA is shown and
to determine in Algorithm
update the 1.
parameters. The specific procedure of the GMCA is shown in Algorithm 1.
Algorithm 1. Gaussian Mixture Clustering Algorithm
Algorithm 1. Gaussian Mixture Clustering Algorithm
Input: Customer data set, the number of Gaussian mixed clusters k, the number of the maximum
AlgorithmInput: Customer
1. Gaussian Mixture dataClustering
set, the number Algorithm of Gaussian mixed clusters k, the number of the maximum
iteration t_max
iteration t_max
Output:Input: Customer
Cluster results data set, the
for each servicenumber of Gaussian mixed clusters k, the number of the maximum
period
Output: Cluster results for each service period
1. iteration t_max
// Initial GMM parameters
1. // Initial GMM parameters
Output: Cluster results for each service period
1. // Initial GMM parameters {(wd, μd, Σd)|1< d < k}←GMM (Ci)
{(wd , µd , Σd )|1< {(wdd,<μk} d, Σd)|1< d < k}←GMM (Ci)
←GMM (Ci )
2. While t< t_max // Iterated clustering
2. While 2. While
t < t_max t< t_max clustering
// Iterated // Iterated clustering
3. For c= 1: n // Calculate posterior distribution (E-step)
3. 3.
For c = 1: n // For c= 1: n //posterior
Calculate Calculatedistribution
posterior distribution
(E-step) (E-step)
4. Calculate posterior probability of each customer to different depots
4. Calculate 4. posterior
Calculate probability
posterior k
of each customer
probability of to
each different
customer depots
to different depots
5. 5.   P ( c |  ,  )   w  P ( c |  ,  ) k
5. cd M d d
 cd d P (c | d , d )   wd  P(c |  d , d )
d d d
1 M
d 1
6. 6. EndEndfor
6. for End for
7. 7. For For
d= 1dto= k1 //toUpdate
k For parameters (M-step)
// Update
7. d= 1 to kparameters (M-step) (M-step)
// Update parameters
8. Calculate
8. the the
Calculate mixing
mixing factor, mean
factor, mean vector,vector,andand covariance
covariance matrixmatrix
8. Calculate
n
the mixing factor, mean vector, and covariance matrix n
9. 9. n n n
w
9.d '  
c 1
 cd / n  d '  
wd ' c 
n
 cd  C c /   cd  d
n
'    cd n
( C c   )  (Cn   )T /  cd
 cd / n c 1 d '    cd c1C c /   cd  d d'  c  cd d (Cc 

n
 )  (Cc   d )T /   cd
1 c 1 d
c 1 c 1 c 1 c 1 c 1
10. 10. Calculate
Calculate thethegap
gapbetween
between thethe updated
updated parameters
parameters and
andthethe
previous parameters
previous parameters
10. Calculate the gap between the updated parameters and the previous paramete
11. 11. If the
If the gapgap is small
is small // Check
// Check the stopthe stop conditions
conditions
11. If the gap is small // Check the stop conditions
12. 12. Update
Update parameters
parameters (wd(w Σd )←d’,(wμdd’, Σ
, μd,, µΣdd,)←(w ’, Σd ’)
µ ’)
12. Update dparameters (wd, μdd, Σd)←(w d’, μd’, Σd’)
13. 13. EndEndif if
13. End if
14. 14. EndEnd for for
14. End for clustering
15. For c= 1: n do //nDetermine
15. For c = 1: do // Determineclustering
16. 15.
Reassign For c=
the customer1: n do the
// Determine clustering
16. Reassign the customer to thetodepot depot
with with the maximum
the maximum probability
probability
17. End 16.
for Reassign the customer to the depot with the maximum probability
17. End for
17. End for

In Algorithm
In Algorithm 1, the
1, the customer
customer clusteringprocedure
clustering procedure includes
includes the
the following
followingseveral
severalsteps:
In Algorithm 1, the customer clustering procedure includes the following sever
First, input the corresponding data of customers, depots, and CSs
steps: First, input the corresponding data of customers, depots, and CSs to construct to construct a data matrix
a
within eachsteps:
serviceFirst, inputand
period, thedefine
corresponding
the number data
of of customers,
mixture depots,
clusters. andinitialize
Second, CSs to construct
data matrix within each service period, and define the number of mixture clusters. Second,
the GMM data matrix within
parameters, eachthe
including service period,
mixing and
factor, define
mean the number
vector, andand of mixturematrix.
covariance clusters. Secon
initialize the GMM parameters, including the mixing factor, mean vector, covariance
Third, initialize
calculate the GMM
the posterior parameters,
probability including
of each the
customer mixing factor,
corresponding mean vector,
to various and covarian
depots.
matrix. Third, calculate the posterior probability of each customer corresponding to vari-
matrix.
Fourth, Fourth,
constantly Third, calculate
update update the
parameters posterior
until the probability
values of each customer corresponding
stable. to va
ous depots. constantly parameters until the of eachof
values parameter remainre-
each parameter
ous depots. Fourth, constantly update parameters until the values of eachthat
parameter r
mainFifth, determine
stable. and classify
Fifth, determine andcustomer
classify clusters
customer based on the
clusters corresponding
based probability
on the corresponding
customers main stable. Fifth, determine and classify customer clusters based on the correspondi
probability thatcan be assigned
customers can beto each depot.
assigned to each depot.
probability that customers can be assigned to each depot.
5.2. Improved Non-Dominated Sorting Genetic Algorithm
NSGA-II, a multi-objective evolutionary algorithm, is widely applied in combinatorial
optimization problems because of its good global search capability [63–65]. In this section, a
heuristic algorithm is applied to address the EVCS-LRPTWRS for simultaneously obtaining
the optimal EV delivery routes and CS selection decisions within the various service periods.
To adapt the algorithm to the proposed EVCS-LRPTWRS in this study, the proposed
INSGA-II has been modified on the basis of the NSGA-II to improve its performance. In the
proposed INSGA-II, the sweep algorithm is introduced to generate a high-quality initial
population, insertion operation is incorporated to select the location of CSs, and an elite
retention mechanism is integrated to avoid prematurely falling into local optima. The
specific procedure of the INSGA-II is shown in Algorithm 2.
Sustainability 2022, 14, 11681 14 of 31

Algorithm 2. INSGA-II
Input: Cluster results, fitness function, population size (pop_size), maximum iterations (r_max)
Output: CS location selection and EV routing plan
1. // Population initialization
2. Generate initial population (Pt ) based on sweep algorithm
3. Calculate the angles between customers and origin
4. Obtain initial solutions by sorting the angles
5. Evaluate the fitness function of the initial population
6. While r ≤ r_max
7. For population = 1: pop_size // Genetic operation
8. Execute tournament selection, order crossover, multi-point mutation to generate
child population (Qt )
9. End for
10. Combine the parent population (Pt ) and child population (Qt ) to generate Rt // Elite
retention
11. If exist that the battery level of EV is negative while arriving customers // Insertion
operation
12. For all routes with negative battery level of EV
13. Perform insertion operation
14. End for
15. End if
16. Repeat // Non-dominated sorting and crowding distance comparison
17. Determine the non-dominated ranking by executing non-dominated sorting
18. Compare the crowding distance of individuals to determine the retention
individuals
19. Until the number of individuals reaches the set population size
20. End

In Algorithm 2, the INSGA-II procedure consists of the following components. First,


generate the initial population based on the sweep algorithm, and evaluate their fitness
value. Second, perform a tournament selection to choose good individuals to crossover and
mutate to produce a child population. Third, perform an insertion operation to make the
delivery routes feasible on the battery level. Fourth, perform a non-dominated sorting and
crowding distance comparison to realize an elite retention mechanism, thus completing the
selection of outstanding individuals. Finally, output the Pareto optimal solution.

5.2.1. Solution Presentation


The presentation of the solution is critical to the efficiency of chromosome-to-route
conversion and affects the performance of the whole algorithm [66]. The natural number
encoding is adopted in this study. Then, the EV delivery routes are obtained through the
EV load capacity constraint. Finally, the appropriate CSs are inserted according to the
power demand to obtain the CS location results, thus realizing the process from encoding to
decoding. In addition, the method of generating the initial solution still affects the efficiency
of obtaining the optimal solution [58]. In the population initialization, the embedding scan
algorithm can improve the convergence speed of the solution and quickly achieve the
optimal solution [67]. In comparison with the traditional random strategy for generating
the initial population, the scanning algorithm sorts the customers by angle calculation to
generate the initial population and obtain the initial individuals with excellent genes.

5.2.2. Genetic Operation


(1) Tournament selection
Each parent chromosome has a certain probability to be selected to produce a child
chromosome, and this probability is related to the fitness value. The chromosome with a
higher fitness and lower objective function has a higher probability of being selected [68,69].
Sustainability 2022, 14, 11681 15 of 31

In this study, the tournament selection method is used to select more suitable chromosomes
for crossover and mutation.
(2) Order crossover
Crossover operators play an important role in genetic algorithms because of their
ability to generate new individuals to improve the search ability of the algorithm [70].
A new individual is generated by replacing and recombining parts of the structure of
two parent individuals [65]. The purpose of the order crossover is to produce new child
chromosomes, and the main execution steps are presented as follows:
Step 1: Randomly select a substring from a parent chromosome.
Step 2: Generate a proto-child by copying the substring into the corresponding position.
Step 3: Delete the genes that are already in the substring from the second parent.
Step 4: Place the genes into the unfixed positions of the proto-child based on the order
of the sequence to generate a child chromosome.
(3) Multi-point mutation
Multi-point mutation is to ensure that the characteristics of the best individuals in
the previous generation are inherited as much as possible in the new individuals of the
next generation [66,67]. The mutation operation is mainly performed by setting multiple
mutation points randomly on the chromosomes of individuals and then performing a
gene segment exchange. In this study, the operator diversifies solutions by randomly
changing the customers’ labeling number in the delivery routes to drive them away from
the local optimum.

5.2.3. CS Insertion Operation


Following the operations of crossover and mutation, some of the current EV distri-
bution routes are not feasible on the battery level of the EVs. The CSs should be inserted
into the infeasible routes to reasonably repair the vehicle routes. The greedy charging
ainability 2022, 14, x FOR PEER REVIEWstrategy is introduced to make the route feasible by identifying the first customer with 17 of
a negative battery level of the EV, and inserting a CS in the part of the route before that
customer [47,60]. The insertion procedure is illustrated in Figure 3.

[1,4 [1,4]
[6,8] C2 [1,4] [6,8] C2 [1,4] CS2
C8 2 1 C7 C8 (+1)1 1 C7
1 D Depot
1 1 CS1
1 [1,4] 1 [1,4]
[6,9] [6,9] C Customer
C3 [0,11] C6 C3 [0,11] C6
1 D 1 [1,4] 1 1 D 1 [1,4]
1 CS Charging satation
[6,10] [6,10]
[6,10] [6,10] 1 [• ,•] Time windows
C4 1 1 C5 C4 1 C5
C1 C1
(a) Infeasible route (b) Improved route after CS insertion

FigureFigure
3. Improved route
3. Improved after
route afterthe
the CS insertion
CS insertion procedure.
procedure.

In Figure 3a, an infeasible route, including eight customers and one depot, is depicted.
In Figure3b,3a,
In Figure S1 an infeasible
is inserted route,
between including
C2 and eight customers
C8 by maintaining and oneofdepot,
both time windows the is d
picted. In Figure
depot and the 3b, S1 is
battery inserted
level between
of the EV C2The
feasibility. and C8 byinsertion
detailed maintaining both
procedure time window
is shown
of theindepot and 3.
Algorithm the battery level of the EV feasibility. The detailed insertion procedure
shown in Algorithm 3.

lgorithm 3 CS insertion procedure


For each infeasible route
Repeat
Identify the first customer with the negative battery level of the EV
Search for all feasible insertion points according to time windows and battery level before the first customer
Sustainability 2022, 14, 11681 16 of 31

Algorithm 3 CS insertion procedure


1. For each infeasible route
Repeat
2. Identify the first customer with the negative battery level of the EV
3. Search for all feasible insertion points according to time windows and battery level
before the first customer
4. Select a CS from the set of candidate CSs to insert
5. For each candidate CS
6. Calculate the sum of distances from CS to the previous customer and the next
customer for each feasible insertion point as the distance increment
7. End for
8. For each feasible insertion point
9. Compare the distance increments between the feasible insertion point and all
candidate CSs
10. Select the CS with the minimum distance increment to insert
11. Calculate and compare the delivery cost increment for the CS insertion
12. End for
13. Determine the final insertion point with the minimum delivery cost increment and
insert the selected CS for the route
14. Until all of the delivery routes are feasible on the battery level of the EV
15. End for

In Algorithm 3, the main idea of the greedy charging strategy consists of determining
the first customer in the infeasible route at which the EV arrives with a negative battery
level and inserting the “best” (which increases the distance least) CS on the route between
that customer and the previous one.

5.2.4. Elite Retention Mechanism


An elite retention mechanism is introduced to merge the parent population with the
child population for the non-dominated sorting, which makes the search space larger [71,72].
Then, individuals with a higher priority are selected after generating the next generation
of populations. Meanwhile, the crowding degree is applied to select individuals at the
Sustainability 2022, 14, x FOR PEER REVIEW 18 of 32
same level, which ensures that the good individuals can have a higher probability of being
retained [63,71]. The specific elite retention mechanism is shown in Figure 4.

Non-dominated Crowding distance


sorting comparation
Pt rank 1 rank 1

Generated by rank 2 rank 2


Pt+1 (N)
sweep
algorithm
rank 3
rank 3
Rt (2N)
Qt

Generated
Eliminated (N)
by genetic
operation
rank k-1
rank k
Figure 4. Elite retention mechanism.
Figure 4. Elite retention mechanism.

In
In Figure
Figure 4,
4, Q
Qtt is
is the
the child
child population,
population, PPtt isis the
the parent
parent population,
population, and
and NN is
is the
the set
set
population size. The elite retention mechanism can
population size. The elite retention mechanism can be realized be realized by the following steps:
Step
Step1:1:Combine
Combine the parent
the population
parent populationwith withthe the
child population
child to form
population a newapop-
to form new
ulation, and and
population, perform
performthe non-dominant sorting
the non-dominant to rank
sorting individuals.
to rank individuals.
Step 2: Place the non-dominant individuals of the different ranks into the new parent
population based on the Pareto rank order
Step 3: Calculate the crowding distance for all individuals of rank k + 1 and arrange
all individuals in a descending order by crowding distance. Eliminate all individuals with
a rank greater than k + 1 if the conditions that the total number of individuals is less than
Sustainability 2022, 14, 11681 17 of 31

Step 2: Place the non-dominant individuals of the different ranks into the new parent
population based on the Pareto rank order
Step 3: Calculate the crowding distance for all individuals of rank k + 1 and arrange
all individuals in a descending order by crowding distance. Eliminate all individuals with
a rank greater than k + 1 if the conditions that the total number of individuals is less than
N after all of the individuals of rank k are placed into the new parent set, and the total
number of individuals is greater than N after all of the individuals of rank k + 1 are placed
into the new parent set, are reached.
Step 4: Add individuals to the new parent population as in steps 2 and 3 until the
number of individuals in the new parent population is equal to N, and eliminate the
remaining individuals.

5.2.5. Elite Retention Mechanism


The non-dominance ranking is a concept based on the Pareto dominance, which facili-
tates the selection of more superior individuals [73,74]. The procedure of non-dominated
sorting consists of three components: First, find all of the individuals in the population that
have not been dominated, and deposit these individuals in Pareto rank 1. Second, deposit
individuals that are dominated only by individuals in rank 1 into Pareto rank 2. Third,
repeat until all individuals have been given a Pareto rank. The set of individuals with
Pareto rank 1 is the Pareto optimal solutions, and the curve formed by these individuals is
the Pareto front.
The crowding distance comparison is adopted to select better individuals from the
same level of solutions [74,75]. The crowding distance is a measure of the density of
individuals on the same Pareto front and can be calculated by using Equation (40)

f n ( j + 1) − f n ( j − 1)
disn ( j) = (40)
f n max − f n min

where j + 1 and j − 1 are two adjacent individuals on the same Pareto front, and f nmax and
f nmin are the maximum and minimum values of the objective function n, respectively.

6. Computational Experiments
In this section, the numerical experiments are presented to verify the proposed solution
method for solving the CS-LRPTWRS. In Section 6.1, an algorithm comparison is conducted
to demonstrate the superiority of the proposed hybrid algorithm in this study. In Section 6.2,
an empirical study is introduced to test the validity and practical implications of the
proposed methodology. In Section 6.3, the management insights of EVCS-LRPTWRS
are summarized.

6.1. Algorithm Comparison


The multi-objective genetic algorithm (MOGA) [76], multi-objective particle swarm
optimization (MOPSO) [77] algorithm, multi-objective ant colony optimization (MOACO)
algorithm [78] have a good performance in solving multi-objective optimization problems.
The three listed algorithms are adopted to compare with the proposed INSGA-II in this
study, to verify its effectiveness and superiority. A set of 20 benchmark instances of the
MDVRPTW collected from the network and emerging optimization research group, are
modified to evaluate the performance of the proposed INSGA-II in solving the EVCS-
LRPTWRS. In the algorithm comparison of the EVCS-LRPTWRS, the locations of the
candidate CSs are generated by randomly selecting from the group of customers, and the
service periods are determined based on the customer’s time windows. The characteristics
of the modified instances, including the number of depots, the number of customers, the
number of candidate CSs, the number of service periods, and the EV loading capacity are
illustrated in Table 5.
Sustainability 2022, 14, 11681 18 of 31

Table 5. Characteristics of the tested dataset.

Number Number of Number Number EV


Instances Datasets
of Depots Customers of CSs of Periods Capacity
1 Pr01_15CS_4W 4 48 15 4 200
2 Pr02_15CS_4W 4 96 15 4 195
3 Pr03_15CS_4W 4 144 15 4 190
4 Pr04_15CS_4W 4 192 15 4 185
5 Pr05_15CS_4W 4 240 15 4 180
6 Pr06_15CS_4W 4 288 15 4 175
7 Pr07_20CS_6W 6 72 20 6 200
8 Pr08_20CS_6W 6 144 20 6 190
9 Pr09_20CS_6W 6 216 20 6 180
10 Pr10_20CS_6W 6 288 20 6 170
11 Pr11_15CS_4W 4 48 15 4 200
12 Pr12_15CS_4W 4 96 15 4 195
13 Pr13_15CS_4W 4 144 15 4 190
14 Pr14_15CS_4W 4 192 15 4 185
15 Pr15_15CS_4W 4 240 15 4 180
16 Pr16_15CS_4W 4 288 15 4 175
17 Pr17_20CS_6W 6 72 20 6 200
18 Pr18_20CS_6W 6 144 20 6 190
19 Pr19_20CS_6W 6 216 20 6 180
20 Pr20_20CS_6W 6 288 20 6 170

The proposed INSGA-II is compared with the MOGA, MOPSO, and MOACO to
solve the EVCS-LRPTWRS based on the datasets given in Table 5. The parameters used
in the MOGA and INSGA-II are set as follows: population size pop_size = 100, selection
probability ps = 0.5, crossover probability pc = 0.9, and mutation probability pm = 0.1. The
key parameters of the MOPSO are as follows: inertia weight w = 0.7, personal learning
coefficient g1 = 1, and global learning coefficient g2 = 2. The parameters of the MOACO
are as follows: pheromone importance a = 5, heuristic factor importance b = 5, pheromone
evaporation coefficient c =0.5. In addition, the maximum iteration number of four algo-
rithms is set as g_max = 500. The comparison results of the three algorithms containing the
TOC, the number of used EVs, and the number of selected CSs are shown in Table 6. The
optimization solutions (i.e., the TOC, the number of EVs, and the number of selected CSs)
of each instance corresponding to the three algorithms are obtained from the Pareto fronts,
and the different Pareto fronts can be evaluated by the bi-objective function value.
In Table 6, the reasonableness of the algorithm comparison is demonstrated by the
significant difference between the computation results of the t-test and p-value. Based on
the average values, the costs of the four algorithms are USD 16,169, USD 17,101, USD 16,567,
and USD 17,279, and the numbers of the selected CSs are 9, 11, 10, and 12, respectively,
indicating that the proposed algorithm performs better in terms of cost and number of
selected CSs. According to the calculation results of each set of instances, the optimized
number of used EVs corresponding to the INSGA-II is basically smaller than the number
of used EVs corresponding to the other three algorithms. This notion means that the
MOPSO, MOGA, and MOACO are inferior to the performance of the INSGA-II. Therefore,
the INSGA-II proposed in this study is effective and has a better performance in the
calculation of the large and medium-sized multi-objective EV routing optimization and CS
location problems.
Sustainability 2022, 14, 11681 19 of 31

Table 6. Algorithm comparison results.

INSGA-II MOGA MOPSO MOACO


Instance
TOC (USD) EVs CSs TOC (USD) EVs CSs TOC (USD) EVs CSs TOC (USD) EVs CSs
1 4953 3 4 5785 4 7 5355 4 4 6075 4 7
2 9062 7 6 10,051 8 9 9072 7 7 10,141 8 9
3 12,774 10 8 13,574 12 10 13,061 10 10 13,696 12 11
4 17,531 15 11 18,357 17 11 17,597 15 11 18,479 17 12
5 21,796 20 14 22,993 20 14 21,975 21 16 23,218 20 15
6 25,687 23 15 26,776 25 15 26,560 24 15 26,878 26 15
7 6931 4 3 8048 5 4 7499 4 3 8162 6 4
8 13,590 6 7 14,416 8 10 13,595 6 9 14,662 8 11
9 21,792 15 9 22,849 16 9 22,346 16 11 22,905 17 10
10 26,795 22 16 27,831 24 18 27,417 22 18 27,939 24 19
11 5062 4 4 5994 4 5 5598 5 4 6354 5 6
12 9278 7 7 9855 7 9 9554 7 7 10,101 8 10
13 12,377 11 8 13,081 12 8 12,548 12 9 13,407 12 10
14 17,942 14 11 18,614 14 12 18,581 14 11 18,680 14 13
15 22,188 20 13 23,190 20 13 22,715 20 15 23,555 21 14
16 25,725 22 15 26,794 23 16 25,726 23 15 26,970 24 16
17 7274 4 4 8208 5 5 7691 4 4 8269 6 6
18 13,579 7 8 14,769 9 10 14,076 8 9 14,991 9 12
19 22,193 15 10 23,308 15 13 22,257 15 10 23,479 15 14
20 26,841 21 15 27,527 21 16 27,126 22 16 27,628 22 16
Average 16,169 13 9 17,101 13 11 16,567 13 10 17,279 14 12
t-test −22.75 −5.41 −5.15 −6.41 −3.94 −4.56 −24.71 −7.63 −7.76
p-value 1.51 × 10−15 1.35 × 10−5 2.84 × 10−5 1.89 × 10−6 4.37 × 10−4 3.83 × 10−4 3.30 × 10−16 1.69 × 10−7 1.30 × 10−7
Sustainability 2022, 14, x FOR PEER REVIEW 2
Sustainability 2022, 14, 11681 20 of 31

6.2. Empirical Study


6.2. Empirical Study
6.2.1. Data Source
6.2.1. Data Source
An empirical
An empirical study based onstudy based onEV
a real-world a real-world
distributionEV distribution
network network in Chon
in Chongqing
City, China is employed to is
City, China test the applicability
employed and
to test the effectiveness
applicability ofeffectiveness
and the proposedofsolution
the proposed so
methods (i.e., bi-objective nonlinear
methods (i.e., programming
bi-objective nonlinearmodel, hybrid heuristic
programming algorithm,
model, hybrid and algorithm
heuristic
resource sharing strategy) in solving the EVCS-LRPTWRS. The EV distribution network
resource sharing strategy) in solving the EVCS-LRPTWRS. The EV distribution is ne
described in Figure 5 and Table
is described 7.
in Figure 5 and Table 7.

C3 C2 C1 C101 C100 C64


CS11
C60 C111
C63
C148 C151 C127 C143 C62 C65 C159
C108
C12 C22 C28 C59 CS3 C61 C160 C107 C110
C7 C144 C27
D3 C145 C68
C5
CS1 C141 C58 C57 C29 C109 C55
C23 C66
C11 C149 C99 C146 CS4
C161
C13 C9 C10 D1 C102 C86 C56 C106 C87
C142 CS2 C54
C4 C20 C24
C150 D2 C25 C103 C30 C50 C153
C153
C128 C26 C67 C113
C17 C152 C124 CS6
C16 C15 C14 CS7 C105
C52
C31
C88 C126 C51
CS5
C147 C21 C104
C40 C93 C112

C36 C19 C131 C32 CS15 C89 C125 C84


C18 C92 C154
C123
C37 C34 C129 C91 C33 C122
CS14 C130 C49 C137 CS9 C85
C139 C132 C98 C71 C69
C39 C138 C41 C70 C155 C83
C37 D5 C121 C114
C157 C119
C6 C94 C97 C48
C140 CS10 C116 C156 C80
C38
CS13 C95 C96 C45 C72 C73 C158
D4 C74
C81
C42 C136 C118 C115 C77 CS8
C8 C133 C43 C120
C134 C46 C82
C44 C135 C47 CS12 C76 C75 C78 C79 C90
C117

Depot Customer Candidate CS

Figure 5. Spatial Figure 5. Spatial


distribution distribution
of the of the logistics
logistics network network in the EVCS-LRPTWRS.
in the EVCS-LRPTWRS.

Table 7. Initial characteristics of characteristics


Table 7. Initial the logistics network in the EVCS-LRPTWRS.
of the logistics network in the EVCS-LRPTWRS.

1st Period
1st Period 2st Period 2st Period
Depots Customers Number of
Depots Number
Customersof Services Time Number of Services
Time
Time
Number of Services Time
Customers Services
Windows Customers Windows
Windows Customers Windo
D1 C1-C30 17 Customers
[0,10] 13 [12,24]
D2 C31-C65
D1 C1-C3019 17 [0,10] [0,10] 16 13[12,24] [12,24
D3 C66-C98
D2 C31-C65 15 19 [0,10] [0,10] 18 16[12,24] [12,24
D4 C99-C132 20 [0,10] 14 [12,24]
D5 D3
C133-C161 C66-C98 13 15 [0,10] [0,10] 16 18[12,24] [12,24
D4 C99-C132 20 [0,10] 14 [12,24
Total 84 77
D5 C133-C161 13 [0,10] 16 [12,24
Total 84 77
In Figure 5 and Table 7, the EV distribution network is composed of five depots (D1,
D2, D3, D4, and D5), a set of 15 candidate CSs (CS1, CS2,..., CS15), and 161 customers
In Figure 5 and Table 7, the EV distribution network is composed of five depo
(C1, C2,..., C161). The depots are marked as black stars and have various service periods.
D2, D3, D4, and D5), a set of 15 candidate CSs (CS1, CS2,..., CS15), and 161 custome
The customers are served by the different logistic facilities in various service periods, and
C2,..., C161). The depots are marked as black stars and have various service period
the candidate CSs can be simultaneously used by multiple EVs. The parameters used
customers are
in the model formulation of served by the differentthat
the EVCS-LRPTWRS logistic
can facilities in various
be set based on theservice periods, a
related
candidate CSs can be simultaneously
references [12,42] and actual surveys, are shown in Table 8.used by multiple EVs. The parameters used
model formulation of the EVCS-LRPTWRS that can be set based on the related refe
[12,42] and actual surveys, are shown in Table 8.
FRr Unit operating cost of CS 50
v Charging rate of EV v, v∈V, (unit: kWh/h) 25
d Variable cost coefficient of depot d, d∈D 0.4
Sustainability 2022, 14, 11681 MNv Annual maintenance cost of EV v, v∈V 21 of200
31
K Number of working periods in one year 52
t Arc specific coefficient of ET t, t∈T 0.13
Table 8. Parameter settings of the model formulation in the EVCS-LRPTWRS.
t Vehicle specific coefficient of ET t, t∈T 0.79
Parameter Description αv Arc specific coefficient of EV v, v∈V Value 0.11
fe β Vehicle
Electricity price, (unit: USD/kWh)
v specific coefficient of EV v, v∈V 1.8 0.76
Fcc CC of electricity per unitDriving
time, (unit: USD/h)
speed of EV v in the nth route within the wth service 4.5
pe-
w
FRr Unit operatingSV cost
vn of CS 50 11.11
δv Charging rate of EV v, v∈riod, v∈V,
V, (unit: n∈ N vw , w∈W, (unit: m/s)
kWh/h) 25
µd Variable cost coefficient of Driving d∈D of EV v in the mth route within the wth service pe-
depot d, speed 0.4
MNv STtmw cost of EV v, v∈V
Annual maintenance 200 11.11
K Number of working periods riod, in t∈T, m∈ M t , w∈W, (unit: m/s)
one year
w
52
λt ET t, t∈Tof EV v in the nth route within the wth service period, v∈V,
Arc specific coefficient ofWeight 0.13
w
γt WEV
Vehicle specific coefficient
vn of ETw t, t∈T 0.79 1910
αv
n∈
Arc specific coefficient of EVNv,v v∈V
, w∈W, v∈V, (unit: kg) 0.11
βv Vehicle specific coefficient of EV v, v ∈ V
Weight of ET t in the mth route within the wth service period, t∈T, 0.76
w
SVvnw Driving speedWETof EVtm v in the nth
w
w route within the wth service period, v∈V, n∈ Ny , w∈W, (unit: m/s) 11.11 4000
w m∈ M t , w∈W, v∈V, (unit: kg)
STtm Driving speed of EV v in the mth route within the wth service period, t∈T, m∈ Mtw w∈W, (unit: m/s) 11.11
WEVvn w Weight of EV vQT in the
t nthLoad capacity
route within of ET
the wth service period, v∈V, n∈ Nyw , w∈W, v∈V, (unit: kg)
t, t∈T 1910 1000
WETtm w Weight of ET tQVin the mth route within the wth service period, t ∈ T, m ∈ M w , w∈W, v∈V, (unit: kg) 4000
v Load capacity of EV v, v∈V t 600
QTt Load capacity of ET t, t∈T 1000
QVv Load capacity ofPEVe PC
v, v∈ V per unit of time of earliness 600 4
Pe Pl of earliness
PC per unit of time PC per unit of time of delay 4 8
Pl PC per unit of time
Ev of delay Battery capacity of EV v, v∈V, (unit: kWh) 8 48
Ev Battery capacity of EV v, v∈V, (unit: kWh) 48

6.2.2. Optimization Results


6.2.2. Optimization Results
(1) Customer clustering results
(1) Customer clustering results
Customers are clustered by the Gaussian hybrid clustering algorithm according to
Customers are clustered by the Gaussian hybrid clustering algorithm according to
their coordinates and time windows in order to simplify the computational complexity of
their coordinates and time windows in order to simplify the computational complexity
the EVCS-LRPTWRS and to determine the number of service periods. Every customer can
of the EVCS-LRPTWRS and to determine the number of service periods. Every customer
be reassigned to the depot
can be reassigned to the service with awith
depot service certain probability
a certain withwith
probability the the
cooperation of of
cooperation mul-
tiple multiple
depots and
depots and customer service sharing. The clustering results are shown in Figure 6.6.
customer service sharing. The clustering results are shown in Figure

Customers served by D1
Customers served by D2
18 Customers served by D3
Customers served by D4
16
Customers served by D5
14
2nd period
12
Time

1st period
10
8
6
29.
60
4 29.
58
48

29.
6.

56
10

de
52

29.
itu
6.

Lat 54
10

ng
56

it ude 29.
6.

Lo
10

52
60
6.
10

Figure 6. Customer clustering results of the EVCS-LRPTWRS.


Figure 6. Customer clustering results of the EVCS-LRPTWRS.
In Figure 6, customers with similar characteristics are divided into a cluster. A cus-
tomer will be assigned to a depot based on the geographic coordinates and time windows
of the customer. For example, customers assigned to D3 are represented by blue balls.
Meanwhile, customers assigned to D4 are represented by green balls. These customers are
classified into two tiers. The lower and upper layers represent the customers to be served
in the first and second service periods, respectively. In summary, different customers will
be served by the corresponding depots in different service periods.
In Figure 6, customers with similar characteristics are divided into a cluster. A
tomer will be assigned to a depot based on the geographic coordinates and time win
of the customer. For example, customers assigned to D3 are represented by blue
Meanwhile, customers assigned to D4 are represented by green balls. These custome
Sustainability 2022, 14, 11681 22 of 31
classified into two tiers. The lower and upper layers represent the customers to be se
in the first and second service periods, respectively. In summary, different customer
be served by the corresponding depots in different service periods.
(2) CS location(2)andCS thelocation
routingand
optimization
the routingresults
optimization results
Based on the previous customer
Based on the previous customer clustering results, clustering
the CSresults, thestrategy
selection CS selection
in eachstrategy in
service period is performed by selecting the appropriate CSs for all
service period is performed by selecting the appropriate CSs for all of the delivery routes of the delivery r
that need to be inserted CSs for EVs. Table 9 and Figure 7 show
that need to be inserted CSs for EVs. Table 9 and Figure 7 show the CS location selection the CS location sele
strategies for the two
strategies for the two service periods. service periods.

Table
Table 9. CS location 9. CS location
selection selection
results of results
the various of the various depots.
depots.

Depots Depots 1st 1st 2nd 2nd CS Location


CS Location
D1 D1 CS1 CS1 CS1 CS1
D2 D2 CS13, CS14
CS13, CS14 CS1, CS14
CS1, CS14 CS1, CS3, CS1,
CS4 CS3, CS4
D3 D3 CS3, CS4 CS3, CS4 CS3, C6CS3, C6 CS6, CS8, CS6, CS8, CS9, CS10
CS9, CS10
D4 D4 CS6, CS9, CS12
CS6, CS9, CS12 CS8, CS10
CS8, CS10 CS12, CS13, CS14
CS12, CS13, CS14
D5 CS10, CS13 CS10
D5 CS10, CS13 CS10

C3 C2 C1 C101 C100 C64


CS11
C60 C111
C63
C148 C151 C127 C143 C62 C65 C159
C108
C12 C22 C28 C59 CS3 C61 C160 C107 C110
C7 C144 C27
D3 C145 C68
C5
CS1 C141 C58 C57 C29 C109 C55
C23 C66
C11 C149 C99 C146 CS4
C161
C13 C9 C10 D1 C102 C86 C56 C106 C87
C142 CS2 C54
C4 C20 C24
C150 D2 C25 C103 C30 C50 C153
C153
C128 C26 C67 C113
C17 C152 C124 CS6
C16 C15 C14 CS7 C105
C52
C31
C88 C126 C51
CS5
C147 C21 C104
C40 C93 C112

C19 C131 C32 CS15 C84


C36 C18 C92 C89 C154C125
C123
C37 C34 C129 C91 C33 C122
CS14 C130 C49 C137 CS9 C85
C139 C132 C98 C71 C69 C155 C83
C39 C138 C41 C70
C37 D5 C121 C114
C157 C119
C6 C94 C97 C48
C140 CS10 C116 C156 C80
C38
CS13 C95 C96 C45 C72 C73 C158
D4 C74
C81
C136 C118 C115 C77 CS8
C8 C133 C43 C42 C120
C134 C46 C82
C44 C135 C47 CS12 C76 C75 C78 C79 C90
C117

Unselected CS Selected CS

Figure
Figure 7. CS location 7. CS location
selection results. selection results.

In Table 9 and Figure 7, 9a and


In Table totalFigure
of ten 7,
CSs are selected
a total of ten CSs forare
charging
selectedbyforthe EVs, namely,
charging by the EVs, na
CS1, CS3, CS4, CS6,
CS1, CS8, CS9, CS6,
CS3, CS4, CS10,CS8,
CS12,CS9,CS13,
CS10,and CS14.
CS12, In the
CS13, andfirst service
CS14. In theperiod, all
first service perio
nine CSs are used by the EVs. Meanwhile, only six CSs are used by the EVs in the
nine CSs are used by the EVs. Meanwhile, only six CSs are used by the EVs in the se second
service period. service
Some CSs are shared
period. Some CSsin different
are shared depots and service
in different depotsperiods in theperiods
and service same in the
depot. For example,
depot.CS1
Forisexample,
shared by CS1D1isin periods
shared by 1D1andin 2, and CS10
periods 1 and is 2,
shared between
and CS10 is shared bet
D4 and D5 in period
D4 and 2. D5
Table 10 and2.Figure
in period Table 810summarize
and Figurethe optimization
8 summarize theresults of the results o
optimization
EVCS-LRPTWRS, including the electricity
EVCS-LRPTWRS, includingconsumption
the electricity(EC), electric delivery
consumption cost (EDC),
(EC), electric delivery cost (E
charging cost (CC), penalty
charging costcost (PC),
(CC), centralized
penalty transportation
cost (PC), cost (CTC), thecost
centralized transportation operating
(CTC), the oper
cost of depots (DOC), the operating cost of CSs (COC), the total operating cost (TOC), the
number of used EVs (Used EVs), the number of shared routes (Shared routes), and the
number of selected CS locations (Used CSs).
Sustainability 2022, 14, x FOR PEER REVIEW 24 of 32

Sustainability 2022, 14, 11681


cost of depots (DOC), the operating cost of CSs (COC), the total operating cost (TOC), 23
theof 31
number of used EVs (Used EVs), the number of shared routes (Shared routes), and the
number of selected CS locations (Used CSs).
Table
Table 10.10. Optimization
Optimization results
results of of
thethe EVCS-LRPTWRS.
EVCS-LRPTWRS.

Service Pe- EC EC EDC


Service EDC CC CC PC PCCTC CTCDOCDOCCOCCOC TOC TOC UsedUsedShared
Shared Used
Used
CaseCase ETs
ETs
Period(kWh)
riod (kWh)(USD) (USD) (USD)
(USD) (USD)(USD)
(USD)(USD)
(USD) (USD) (USD)(USD)
(USD) (USD) EVsEVs Routes
Routes CSs
CSs
1st 1st 35693569 64246424 197 197 286 286 0 0 500 500 0 0 10,976 10,976 19 19 0 0 00 8 8
Initial
Initial 2nd 2nd 27702770 49864986 108 108 154 154 0 0 500 500 0 0 85188518 13 13 0 0 00 7 7
TotalTotal 63396339 11,410
11,410 305 305 440 440 0 0 1000 1000 0 0 19,494 19,494 32 32 0 0 00 1515
1st 1st 26532653 47754775 162 162 107 107 237 237 536 536 300 300 87708770 9 9 7 7 11 9 9
Opti-
2nd
Optimized 2nd 21782178 39203920 83 83 96 96 186 186 513 513 200 200 71767176 5 5 5 5 1 1 7 7
mized
TotalTotal 48534853 86958695 245 245 203 203 423 423 1049 1049 500 500 15,968 15,968 9 9 12 12 11 1010

Initial logistics network


18 7500 Optimized logistics network 500 35 14
24,000
32
16 440 12
6339 15 30 12
20,000 19,494
6000 400
Electricity consumption (kWh)

14
Number of selected CSs

Total operating Cost (USD)

10

Number of shared EVs


25

Number of used EVs


15,968

Penalty Cost (USD)


12 16,000 4853
4500 10 300 8
10 20
12,000
8 6
203 15
3000 200
6 8000 4
9 10
4 1500 100 2
4000 5
2 0
0
0 0 0 0 0
h)

s
)
SD

SD

te

CS
EV
kW

u
U

ro

d
d
(

C(

se
se
EC

PC

de

U
U
TO

ar
Sh

Figure
Figure 8. Optimization
8. Optimization results
results of EVCS-LRPTWRS.
of EVCS-LRPTWRS.

In In Table
Table 1010andand Figure
Figure 8,8,
thetheEC,
EC,PC,
PC,TOC,
TOC,thethenumber
numberof ofselected
selectedCSs,
CSs,and
and the
the num-
number
berofofused
used EVs in both
EVs twotwo
in both service periods
service present
periods a downward
present a downwardtrend compared with thewith
trend compared initial
logistics network. The TOC is reduced from USD 19,494 to USD 15,968
the initial logistics network. The TOC is reduced from USD 19,494 to USD 15,968 with the with the vehicle
routing
vehicle optimization,
routing a decrease
optimization, of 18%.
a decrease The The
of 18%. EC dropped
EC dropped fromfrom
63396339
kWh to 4853
kWh kWh,
to 4853
thereby obtaining 1986 kWh of electricity savings. The PCs decreased
kWh, thereby obtaining 1986 kWh of electricity savings. The PCs decreased from USD 440 from USD 440 to USD
203, indicating that the violations of customers’ time windows are alleviated.
to USD 203, indicating that the violations of customers’ time windows are alleviated. Fur- Furthermore,
the number
thermore, of selected
the number CSs dropped
of selected from 15 tofrom
CSs dropped 10 and the10number
15 to and theofnumber
used EVsof shifted
used EVs from
32 to from
shifted 9 with32the
toimplementation of the transportation
9 with the implementation resource sharing
of the transportation strategy.
resource The delivery
sharing strat-
egy. The delivery routes from the depots to their service customers can be optimized andby
routes from the depots to their service customers can be optimized and reconstructed
the proposed
reconstructed byINSGA-II described
the proposed in Section
INSGA-II 5.2. The
described further5.2.
in Section optimization
The furtherresults of the
optimiza-
scheme of the delivery routes and the transportation resource sharing strategies for the
tion results of the scheme of the delivery routes and the transportation resource sharing
EVCS-LRPTWRS are shown in Table 11 and Figure 9.
strategies for the EVCS-LRPTWRS are shown in Table 11 and Figure 9.
In Table 11 and Figure 9, nine EVs perform a total of 16 routes for delivery tasks.
Eleven delivery routes are executed by nine EVs in the first service period, and five de-
livery routes are completed by three EVs in the second service period. Each EV can be
shared within a service period and among several service periods with the implementation
of the transportation resource sharing strategy. For example, EV1 initially serves nine
customers and then serves four customers for the same depot in the first service period,
and it continues to complete one delivery route for D4 in the second service period. The
start and end times of the three shared routes do not overlap. Therefore, EV1 executes
a total of three delivery routes and serves 26 customers, thereby realizing the sharing of
transportation resources and greatly improving the utilization of EVs.
Table 11. Optimized routes of EVCS-LRPTWRS.

Service
Depot 2022,
Sustainability EV 14, 11681 Route 24 of 31
Period
EV1 1st C24→C152→C26→C104→C31→C105→C103→C25→C142
EV2 1st C27→C100→C101→C1→C28→C2→C3→C151→C12→CS1→C9→C10→C99→C144→C146
D1 Table 11. Optimized routes of EVCS-LRPTWRS.
EV1 1st C58→C59→C143→C141
Depot EVEV4Service2nd
Period C21→C112→C14→C20→C23→CS1→C22→C127
Route
EV1EV3 1st C4→C11→C149→C5→C148→C7
1st C24→C152→C26→C104→C31→C105→C103→C25→C142
D2 EV2EV4 1st C27→C100→C101→C1→C28→C2→C3→C151→C12→CS1→C9→C10→C99→C144→C146
1st C34→C36→C39→CS13→C38→C8→C6→C37→CS14→C40→C147→C150
D1 EV1 1st C58→C59→C143→C141
EV4EV5 2nd C15→C93→CS13→C134→CS14→C17→C16→C128→C13
2nd C21→C112→C14→C20→C23→CS1→C22→C127
EV3EV3 1st C106→C153→C67→C161→C66→C68→CS4→C159→C65→C110→C145
1st C4→C11→C149→C5→C148→C7
D3
D2 EV4EV7 1st C107→C64→C62→C108→C109→C61→C29→CS3→C30→C102→C60→CS3→C63→C160
1st C34→C36→C39→CS13→C38→C8→C6→C37→CS14→C40→C147→C150
EV5 2nd C15→C93→CS13→C134→CS14→C17→C16→C128→C13
EV8 2nd C54→C87→C113→C53→CS6→C124→C50→C86→CS3→C57→C111→C56→C55
EV3 1st C106→C153→C67→C161→C66→C68→CS4→C159→C65→C110→C145
EV7
EV9 1st C158→C73→C157→C72→C71→C137→C70→CS9→C69→C154→C155→C156→C74
1st C107→C64→C62→C108→C109→C61→C29→CS3→C30→C102→C60→CS3→C63→C160
D3
EV8 2nd C119→C125→C52→C51→C126→CS6→C88→C89→C121→CS12→C76→C118→C116→C11
C54→C87→C113→C53→CS6→C124→C50→C86→CS3→C57→C111→C56→C55
D4 EV8 1st
EV9 1st 5→C117→C75
C158→C73→C157→C72→C71→C137→C70→CS9→C69→C154→C155→C156→C74
D4 EV8EV1 1st C119→C125→C52→C51→C126→CS6→C88→C89→C121→CS12→C76→C118→C116→C115→C117→C75
2nd C120→C77→C79→C90→C82→CS8→C81→C114→C85→C123→C84→C122→C83→C80
EV1 2nd C120→C77→C79→C90→C82→CS8→C81→C114→C85→C123→C84→C122→C83→C80
EV5 1st C43→C44→C42→C135→C136→C140→CS10→C96→C139
EV5 1st C43→C44→C42→C135→C136→C140→CS10→C96→C139
D5
D5 EV6EV6 1st C94→C133→C138→CS13→C35→C32→C33→C98→C97→C41→C95
1st C94→C133→C138→CS13→C35→C32→C33→C98→C97→C41→C95
EV3EV3 2nd C46→C47→C45→C48→CS10→C49→C92→C132→C91→C131→C19→C129→C18→C130
2nd C46→C47→C45→C48→CS10→C49→C92→C132→C91→C131→C19→C129→C18→C130
CS:Charging
CS: Charging stations
stations selected
selected inroutes.
in the the routes.

Depot
Customer
Selected CS location
Unselected CS location
Route performed by EV1
Route performed by EV2
Route performed by EV3
Route performed by EV4
Route performed by EV5
Route performed by EV6
Route performed by EV7
Route performed by EV8
Route performed by EV9

Figure 9. Optimized routes of the EVCS-LRPTWRS.


Figure 9. Optimized routes of the EVCS-LRPTWRS.
6.2.3. Analysis and Discussion
In Table 11 and Figure 9, nine EVs perform a total of 16 routes for delivery tasks.
(1) Comparison of the different service periods
Eleven delivery routes are executed by nine EVs in the first service period, and five de-
liveryThe determination
routes of the
are completed bydifferent
three EVsnumber
in the of service
second periods
service has an
period. impact
Each on the
EV can be
effect of optimization for EV distribution networks. This section demonstrates
shared within a service period and among several service periods with the implementa- the applica-
bility
tion ofofthe
thetransportation
service periods in the EVCS-LRPTWRS
resource sharing strategy. optimization
For example, byEV1 further analyzing
initially the
serves nine
determination
customers andofthen service periods.
serves The optimization
four customers resultsdepot
for the same underindifferent
the first values
serviceofperiod,
W are
compared to determine
and it continues the optimal
to complete numberroute
one delivery of service periods.
for D4 The comparison
in the second resultsThe
service period. are
shown in Table 12 and Figure 10.
start and end times of the three shared routes do not overlap. Therefore, EV1 executes a
total The optimization
of three deliveryresults,
routes including
and serves the26EC, TOC, the thereby
customers, number realizing
of EVs, andthethe number
sharing of
of selected CS locations under various service period segmentation
transportation resources and greatly improving the utilization of EVs. scenarios are shown in
Table 12 and Figure 10. Compared with the EC of 5921kWh, the TOC of USD 20,050, 18 EVs
in demand, and 14 selected CSs without period segmentation, the corresponding indicators
for the other three scenarios with a service period segmentation are significantly reduced.
When W = 2, the minimum the TOC can reach USD 15,968, which is reduced by USD 770
and USD 2286 compared with those of W = 3 and W = 4, and no significant gap is observed
in the number of EVs and selected CSs. Therefore, dividing the entire work period into
6.2.3. Analysis and Discussion
(1) Comparison of the different service periods
The determination of the different number of service periods has an impact on the
Sustainability 2022, 14, 11681 effect of optimization for EV distribution networks. This section demonstrates the applica-
25 of 31
bility of the service periods in the EVCS-LRPTWRS optimization by further analyzing the
determination of service periods. The optimization results under different values of W are
compared
two servicetoperiods
determine the
in the optimalstudy
empirical number of optimal
is the service choice
periods. Thethe
from comparison
perspectiveresults
of the
are shown in Table 12 and Figure 10.
bi-objective of minimizing the cost and number of vehicles.
Table 12. Comparison results of the different service periods.
Table 12. Comparison results of the different service periods.
W=1 W=2 W=3 W=4
Service
TOC W = 1 W
TOC = 2 W
TOC= 3 W
TOC = 4
Periods
Service EC EVs CSs EC EVs CSs EC EVs CSs EC EVs CSs
(USD)
TOC (USD)
TOC (USD)
TOC (USD)
TOC
Periods EC EVs CSs EC EVs CSs EC EVs CSs EC EVs CSs
1st 5921 20,050
(USD) 18 14 2653 (USD) 8770 9 9 1829 (USD)6351 8 11 1537 (USD)
5276 7 7
2nd
1st / /
5921 20,050 18 / /
14 2178
2653 8770 7176 9 5 9 7 1594
1829 6351 5413 8 4 116 1320
1537 52764658 74 76
3rd
2nd / / // /
/ // /
2178 7176 / 5 / 7 / 1483
1594 5413 4974 4 3 66 1176
1320 46584366 44 65
4th
3rd // // /
/ // // // // / / 1483 / 4974/ 3 / 6/ 980 4366
1176 3954 43 55
Total
4th 5921/ 20,050/ 18
/ 14
/ 4853
/ 15,968
/ /9 /10 4906
/ 16,738
/ /9 /11 5013
980 18,254
3954 310 59
Total 5921
CS1, 20,050
CS2, CS3, 18CS4 14 4853 15,968 9 10 4906 16,738 9 11 5013 18,254 10 9
Selected CS1, CS3, CS4 CS1, CS2, CS3, CS4 CS1, CS3, CS4
CS6,CS1,
CS7,CS2,
CS8,CS3,
CS9,CS4CS10,
CS
Selected CS6,
CS1,CS8, CS9,
CS3, CS4CS10 CS6, CS9,
CS1, CS10,
CS2, CS12
CS3, CS4CS13, CS6, CS8,
CS1, CS9,
CS3, CS10
CS4
CS11,
CS6, CS7, CS12,
CS8, CS9, CS10,
locations
CS CS12,
CS6, CS8, CS13,
CS9,CS14
CS10 CS6, CS9,CS14,
CS10,CS15
CS12 CS13, CS6,CS13, CS14CS10
CS8, CS9,
CS13, CS14,
CS11, CS15
CS12,
locations CS12, CS13, CS14 CS14, CS15 CS13, CS14
CS13, CS14, CS15

7000 20 W=1 25,000 16


18 W=2
18 14
6000 5921 W=3 14
20,050 W=4
16 4906 20,000
18,254 12
5
Electricity consumption (kWh)

5000 4853 15,968 11


14

Total operating cost (USD)

Number of selected CSs


15,968
Number of used EVs

10
10
12 9 15,000
4000
10
10 8
9 9
3000
8 10,000
6

2000 6
4
4 5000
1000 2
2

0 0 0 0
EC(kWh) TOC(USD) Used EVs CSs

Figure 10. Comparison results of the different service periods.


Figure 10. Comparison results of the different service periods.
(2) Comparison of different resource sharing strategies
The optimization results, including the EC, TOC, the number of EVs, and the number
Three scenarios of EV sharing are considered based on the EV routing optimization
of selected CS locations under various service period segmentation scenarios are shown
and the CS location selection analysis for two service periods to prove the effectiveness and
in Table 12 and Figure 10. Compared with the EC of 5921kWh, the TOC of USD 20,050, 18
practicality of the transportation resource sharing strategy: (1) non-sharing of EVs in the
EVs in demand, and 14 selected CSs without period segmentation, the corresponding in-
EV distribution network; (2) internal sharing of EVs in the EV distribution network among
dicators for the other three scenarios with a service period segmentation are significantly
various periods in the same depot; and (3) global sharing of EVs in the EV distribution
reduced. When W = 2, the minimum the TOC can reach USD 15,968, which is reduced by
network among various periods or different depots. The comparison results of the number
USD 770 and USD 2286 compared with those of W = 3 and W = 4, and no significant gap
of used EVs, the number of shared routes, the EC, the TOC, and the number of the selected
is observed in the number of EVs and selected CSs. Therefore, dividing the entire work
CSs under the three sharing modes are shown in Table 13 and Figure 11.
period into two service periods in the empirical study is the optimal choice from the per-
In Table 13 and Figure 11, the TOC is decreased from USD 18,753 to USD 17,954 to
spective of the bi-objective of minimizing the cost and number of vehicles.
USD 15,968 as the number of shared routes is increased from the non-sharing of zero to
the internal sharing of two to the global sharing of 12, and the number of used EVs is
declined from 16 to 12 to 9. Moreover, the EC and the number of selected CSs are decreasing
with the deepening of the transportation resource sharing strategy. The minimum number
of used EVs, the maximum number of shared routes, and the lowest TOC in the third
case prove that the global sharing model is the optimal transportation resource sharing
model. Therefore, the transportation resource sharing strategy is conducive to saving
Three scenarios of EV sharing are considered based on the EV routing optimiza
and the CS location selection analysis for two service periods to prove the effective
and practicality of the transportation resource sharing strategy: (1) non-sharing of EV
the EV distribution network; (2) internal sharing of EVs in the EV distribution netw
among various periods in the same depot; and (3) global sharing of EVs in the EV d
Sustainability 2022, 14, 11681 26 of 31
bution network among various periods or different depots. The comparison results o
number of used EVs, the number of shared routes, the EC, the TOC, and the numb
the selected CSs under the three sharing modes are shown in Table 13 and Figure 11
transportation resources and optimizing the resource allocation, thus achieving a reduction
Table
in the TOC and an 13. Comparison
increase in the results of the three
transportation cases of the resource sharing strategy.
efficiency.

Table 13. Comparison


Caseresults of the Period
Service three casesEC the resourceTOC
of (kWh) sharing Used
strategy.
EVs Shared Routes Selected
(USD)
Case Service Period EC (kWh) TOC (USD)
1st Used
2936EVs Shared Routes
9755 10 Selected
0 CSs 10
Non-
1st 2936 2nd
9755 2461
10 8998 0 6 100 7
Non-sharing 2nd
sharing
2461 8998 6 0 70
Total 5397 18,753 16 13
Total 5397 18,753 16 0 13
1st 2722 9406 10 1 10
1st 2722Internal 9406
2nd 10
2426 8548 1 6 10
1 7
Internal sharing 2nd 2426sharing 8548 6 1 7
Total 5148 17,954 12 2 11
Total 5148 17,954 12 2 11
1st 2653 8770 9 7 9
1st 2653Global 8770 9 7 9
Global sharing 2nd 2178sharing
2nd
7176
2178
5
7176 5 5 5
7
7
Total 4853 Total
15,968 4853
9 15,968 12 9 12
10 10

7000 20
Non–sharing 20,000 16 16
18,753 Internal sharing
17,954
18 Global sharing 14
6000
15,968 16 13
16 5397 12
12
5148 15,000 12
Electricity consumption (kWh)

5000

Total operating cost (USD)


14 4853 11

Number of selected CSs


10
Number of used EVs

Number of shared EVs


10
12
12
4000 8
10 10,000 8
9
3000 6
8
4
2000 6
2 5000 4
2
4
1000 0
2 0

0 0 -2 0
h)

D)

es

s
EV

CS
ut
W

US

ro
(k

d
d
C(

te
se
EC

ed

ec
U
TO

ar

l
Se
Sh

Figure 11. Comparison


Figure 11.results of the three
Comparison cases
results of the
of the transportation
three resource sharing
cases of the transportation strategy.
resource sharing strateg

6.3. Management Insights


In Table 13 and Figure 11, the TOC is decreased from USD 18,753 to USD 17,9
A transportation resource
USD 15,968 as thesharing
numberstrategy is incorporated
of shared into the from
routes is increased optimization of the of ze
the non-sharing
CS location-routing problem for a multi-depot multi-period EV distribution network.
the internal sharing of two to the global sharing of 12, and the number The
of used EV
reduction of the
declined from 16 to 12 to 9. Moreover, the EC and the number of selected the
TOC and the improvement of transportation efficiency demonstrate CSs are dec
benefits of addressing the deepening
ing with the EVCS-LRPTWRS. In this section,resource
of the transportation the management insights of
sharing strategy. The minim
this study arenumber
described as follows:
of used EVs, the maximum number of shared routes, and the lowest TOC in
(1) This study
thirdhas
casea significant practical
prove that the global implication foristhe
sharing model theoptimization of logisticsresource s
optimal transportation
networks.ing model. Therefore, the transportation resource sharing strategy isdelivery
The periodic location selections of CSs contribute to promoting conducive to sa
efficiency via the reassignment of customers according to their characteristics in
different service periods. The optimization of the EV delivery routes greatly alleviates
the phenomenon of long-distance transportation and violations of customers’ time
windows. The adoption of a transportation resource sharing strategy in a multi-
depot multi-period EV distribution network can reduce the number of required EVs,
thus facilitating the construction of resource-efficient and environmentally friendly
logistics networks.
(2) The proposed solution method in this study to solve the EVCS-LRPTWRS can provide
a methodological reference for the research of the EVLRP. The bi-objective nonlinear
Sustainability 2022, 14, 11681 27 of 31

programming model of the minimum TOC and the minimum number of required
EVs balances the operation of the logistics network from two conflicting aspects of
economy and efficiency. In addition, the hybrid algorithm, including a customer
clustering algorithm and a heuristic algorithm, is developed to solve the EVCS-
LRPTWRS. The service periods are divided by customer clustering according to
the time window’s characteristics for improving the transportation efficiency. The
integration of the CS insertion operation into the heuristic algorithm allows for the
reasonable CS location, thus, shortening the delivery distance and saving the operating
cost. Therefore, the proposed solution methods contribute to building an economic
and sustainable logistics network and promoting the enterprise’s competitiveness.
(3) The optimization of the EV distribution networks plays a great role in alleviating the
conflicts between humans and the environment. With the proposal of the dual-carbon
goal: carbon peak and carbon neutrality, EVs are widely adopted by enterprises and
logistics companies to construct a sustainable logistics distribution network for coping
with the increasing competition. Meanwhile, many government departments have
enacted a series of policies to promote the popularity and development of EVs to face
the two challenges of environmental degradation and energy scarcity. Furthermore,
new energy technologies (e.g., vehicle-pile cloud interconnection, wireless charging,
and power exchange technology) should be widely promoted to address the technical
limitations and practical application constraints of EVs.

7. Conclusions
This study proposes a problem of the EVCS-LRPTWRS to simultaneously determine
the locations of CSs and optimize the route plans of EVs. A resource sharing strategy
among multiple depots within the various service periods is introduced to improve the
operational efficiency of the logistics networks and rationalize resource configuration. First,
a bi-objective nonlinear programming model is formulated to minimize the TOC and
the number of EVs. Second, a hybrid algorithm combining the GMCA and INSGA-II is
developed to achieve the Pareto optimal solutions of the proposed problem. Third, an
algorithm comparison between the INSGA-II and three other algorithms (e.g., MOGA,
MOPSO, and MOACO) is carried out to test the performance of the proposed algorithm.
Finally, an empirical study of Chongqing City, China is designed to verify the validity and
practicality of the proposed model and algorithm. In addition, the sensitivity analyses
of the service periods and sharing modes are explored and discussed to demonstrate the
effectiveness of the proposed strategies.
The comparison results between the proposed INSGA-II and MOGA, and MOPSO
have demonstrated the superiority of the proposed algorithm in optimizing the EVCS-
LRPTWRS. In a real-world case study, the TOC is reduced by USD12,047, the number of
EVs is decreased by 29, the selected CS locations is minimized from 15 to 10, and the EC
of EVs is dropped by 1348 kWh after the optimization of the initial logistics network. The
effectiveness and practicality of the proposed solution methods in reducing the operating
costs, number of EVs, and EC in the EVCS-LRPTWRS are proven by the experimental
calculation results. The different service period divisions indicate that dividing the whole
working period into two service periods is the optimal period classification. In addition,
three modes of resource sharing strategy are analyzed and discussed, and the comparison
results illustrate that the global sharing of EVs is superior to the other modes. These
findings show that the transportation resource sharing among multiple depots within
various service periods is an economical and efficient strategy to construct sustainable
logistics networks.
This study has significant implications for academic researchers and logistics en-
terprises. The transportation resource sharing strategy and service period division are
incorporated in the selection of CS locations. Future research can be considered in the
following directions: (1) More constraints (i.e., the capacity of CSs, the nonlinearity of charg-
ing speed, and the influence of cargo loading) can be considered to extend the practicality
Sustainability 2022, 14, 11681 28 of 31

of the mathematical model. (2) Pickup and delivery services can be incorporated into future
studies of the EVLRP, and the synchronization and coordination of the two activities can
be deepened. (3) Dynamics and uncertainty of customer demands can be further studied
in the EVCS-LRPTWRS to fit the actual logistics network considerations. (4) The mode of
transportation resource sharing strategy is worth further exploring to optimize the resource
configuration and improve the transportation efficiency in the logistics network.

Author Contributions: Conceptualization, Y.W. and J.Z. (Jingxin Zhou); methodology, Y.W. and
J.Z. (Jingxin Zhou); software, J.Z. (Jingxin Zhou); validation, J.Z. (Jingxin Zhou) and Y.S.; formal
analysis, Y.W. and J.Z. (Jingxin Zhou); investigation, J.Z. (Jiayi Zhe), Y.S. and X.W.; resources, Y.W.;
data curation, Y.W. and J.Z. (Jingxin Zhou); writing—original draft preparation, Y.W. and J.Z. (Jingxin
Zhou); writing—review and editing, Y.W., J.Z. (Jingxin Zhou), Y.S., X.W., J.Z. (Jiayi Zhe) and H.W.;
visualization, J.Z. (Jingxin Zhou), J.Z. (Jiayi Zhe) and Y.S.; supervision, Y.W. and H.W.; project
administration, Y.W.; funding acquisition, Y.W. All authors have read and agreed to the published
version of the manuscript.
Funding: The authors would like to express their sincere appreciation for the valuable comments
made by three anonymous reviewers, which helped us to improve the quality of this paper. This
research is supported by the National Natural Science Foundation of China (71871035), Key Sci-
ence and Technology Research Project of Chongqing Municipal Education Commission (KJZD-
K202000702), Key Project of the Human Social Science of Chongqing Municipal Education Com-
mission (20SKGH079), Chongqing Liuchuang Plan Innovation Project (cx2021038), Team Building
Project for Graduate Tutors in Chongqing (JDDSTD2019008), Chongqing Bayu Scholar Youth Project
(YS2021058), and Research and Innovation Program for Graduate Students in Chongqing (CYS22424).
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Yang, S.Y.; Ning, L.J.; Tong, L.; Shang, P. Integrated electric logistics vehicle recharging station location–routing problem with
mixed backhauls and recharging strategies. Transp. Res. Part C Emerg. Technol. 2022, 140, 103605. [CrossRef]
2. Asadi, S.; Nilashi, M.; Iranmanesh, M.; Ghobakhloo, M.; Samad, S.; Alghamdi, A.; Almulihi, A.; Mohd, S. Drivers and barriers of
electric vehicle usage in Malaysia: A DEMATEL approach. Resour. Conserv. Recycl. 2022, 177, 105965. [CrossRef]
3. Wang, L.; Gao, S.; Wang, K.; Li, T.; Li, L.; Chen, Z.Y. Time-dependent electric vehicle routing problem with time Windows and
path flexibility. J. Adv. Transp. 2020, 2020, 3030197. [CrossRef]
4. Kchaou-Boujelben, M. Charging station location problem: A comprehensive review on models and solution approaches. Transp.
Res. Part C Emerg. Technol. 2021, 132, 103376. [CrossRef]
5. Li, J.Q.; Han, Y.Q.; Duan, P.Y.; Han, Y.Y.; Niu, B.; Li, C.D.; Zheng, Z.X.; Liu, Y.P. Meta-heuristic algorithm for solving vehicle
routing problems with time windows and synchronized visit constraints in prefabricated systems. J. Clean. Prod. 2020, 250, 119464.
[CrossRef]
6. Chen, R.; Liu, X.L.; Mia, L.X.; Yang, P. Electric vehicle tour planning considering range anxiety. Sustainability 2020, 9, 3685.
[CrossRef]
7. Schiffer, M.; Walther, G. The electric location routing problem with time windows and partial recharging. Eur. J. Oper. Res. 2017,
260, 995–1013. [CrossRef]
8. Ran, C.; Zhang, Y.; Yin, Y. Demand response to improve the shared electric vehicle planning: Managerial insights, sustainable
benefits. Appl. Energy 2021, 292, 112863. [CrossRef]
9. Li, J.L.; Liu, Z.B.; Wang, X.F. Public charging station location determination for electric ride-hailing vehicles based on an improved
genetic algorithm. Sustain. Cities Soc. 2021, 74, 103182. [CrossRef]
10. He, P.; Zhang, S.S.; He, C. Impacts of logistics resource sharing on B2C E-commerce companies and customers. Electron. Commer.
Res. Appl. 2019, 34, 100820. [CrossRef]
11. Xu, X.F.; Hao, J.; Zheng, Y. Multi-objective artificial bee colony algorithm for multi-stage resource leveling problem in sharing
logistics network. Comput. Ind. Eng. 2020, 142, 106338. [CrossRef]
12. Wang, Y.; Sun, Y.Y.; Guan, X.Y.; Fan, J.X.; Xu, M.Z.; Wang, H.Z. Two-echelon multi-period location routing problem with shared
transportation resource. Knowl. Based Syst. 2021, 226, 107168. [CrossRef]
13. Wang, Y.; Assogba, K.; Fan, J.X.; Xu, M.Z.; Liu, Y.; Wang, H.Z. Multi-depot green vehicle routing problem with shared transporta-
tion resource: Integration of time-dependent speed and piecewise penalty cost. J. Clean. Prod. 2019, 232, 12–29. [CrossRef]
Sustainability 2022, 14, 11681 29 of 31

14. Erdelić, T.; Carić, T. A Survey on the electric vehicle routing problem: Variants and solution approaches. J. Adv. Transp. 2019,
2019, 5075671. [CrossRef]
15. Ghorbani, E.; Alinaghian, M.; Gharehpetian, G.B.; Mohammadi, S.; Perboli, G. A survey on environmentally friendly vehicle
routing problem and a proposal of its classification. Sustainability 2020, 12, 9079. [CrossRef]
16. Kucukoglu, I.; Dewil, R.; Cattrysse, D. The electric vehicle routing problem and its variations: A literature review. Comput. Ind.
Eng. 2021, 161, 107650. [CrossRef]
17. Dündar, H.; Ömürgönülşen, M.; Soysal, M. A review on sustainable urban vehicle routing. J. Clean. Prod. 2021, 285, 125444.
[CrossRef]
18. Qin, H.; Su, X.X.; Ren, X.; Luo, Z.X. A review on the electric vehicle routing problems: Variants and algorithms. Front. Eng. Manag.
2021, 8, 370–389. [CrossRef]
19. Xiao, Y.Y.; Zhang, Y.; Kaku, I.; Kang, R.; Pan, X. Electric vehicle routing problem: A systematic review and a new comprehensive
model with nonlinear energy recharging and consumption. Renew. Sustain. Energy Rev. 2021, 151, 111567. [CrossRef]
20. Paz, J.C.; Granada-Echeverri, M.; Escobar, J.W. The multi-depot electric vehicle location routing problem with time windows. Int.
J. Ind. Eng. Comput. 2018, 9, 123–136. [CrossRef]
21. Yang, J.; Sun, H. Battery swap station location-routing problem with capacitated electric vehicles. Comput. Oper. Res. 2015, 55,
217–232. [CrossRef]
22. Yidiz, B.; Arslan, O.; Karasan, O.E. A branch and price approach for routing and refueling station location model. Eur. J. Oper. Res.
2016, 248, 815–826. [CrossRef]
23. He, J.; Yang, H.; Tang, T.Q.; Huang, H.J. An optimal charging station location model with the consideration of electric vehicle’s
driving range. Transp. Res. Part C Emerg. Technol. 2018, 86, 641–654. [CrossRef]
24. Zhang, S.; Chen, M.Z.; Zhang, W.Y. A novel location-routing problem in electric vehicle transportation with stochastic demands.
J. Clean. Prod. 2019, 221, 567–581. [CrossRef]
25. Guo, F.; Huang, Z.H.; Huang, W.L. Integrated location and routing planning of electric vehicle service stations based on users’
differentiated perception under a time-sharing leasing mode. J. Clean. Prod. 2020, 277, 123513. [CrossRef]
26. Calik, H.; Oulamara, A.; Prodhon, C.; Salhi, S. The electric location-routing problem with heterogeneous fleet: Formulation and
Benders decomposition approach. Comput. Oper. Res. 2021, 131, 105251. [CrossRef]
27. Wang, C.L.; Guo, C.C.; Zuo, X.Q. Solving multi-depot electric vehicle scheduling problem by column generation and genetic
algorithm. Appl. Soft Comput. 2021, 112, 107774. [CrossRef]
28. Sadati, M.E.H.; Catay, B. A hybrid variable neighborhood search approach for the multi-depot green vehicle routing problem.
Transp. Res. Part E Logist. Transp. Rev. 2021, 149, 102293. [CrossRef]
29. Lam, E.; Desaulniers, G.; Stuckey, P.J. Branch-and-cut-and-price for the electric vehicle routing problem with time windows,
piecewise-linear recharging and capacitated recharging stations. Comput. Oper. Res. 2022, 145, 105870. [CrossRef]
30. Raeesi, R.; Zografos, K.G. The electric vehicle routing problem with time windows and synchronised mobile battery swapping.
Transportation Res. Part B Methodol. 2020, 140, 101–129. [CrossRef]
31. Li, S.Y.; Huang, Y.X.; Mason, S.J. A multi-period optimization model for the deployment of public electric vehicle charging
stations on network. Transp. Res. Part C Emerg. Technol. 2016, 65, 128–143. [CrossRef]
32. Neves-Moreira, F.; Amorim-Lopes, M.; Amorim, P. The multi-period vehicle routing problem with refueling decisions: Traveling
further to decrease fuel cost. Transp. Res. Part E Logist. Transp. Rev. 2017, 133, 101817. [CrossRef]
33. Lin, B.; Ghaddar, B.; Nathwani, J. Electric vehicle routing with charging/discharging under time-variant electricity prices. Transp.
Res. Part C Emerg. Technol. 2021, 130, 103285. [CrossRef]
34. Brandstätter, G.; Kahr, M.; Leitner, M. Determining optimal locations for charging stations of electric car-sharing systems under
stochastic demand. Transp. Res. Part B Methodol. 2017, 104, 17–35. [CrossRef]
35. Friedrich, M.; Noekel, K. Modeling intermodal networks with public transport and vehicle sharing systems. EURO J. Transp.
Logist. 2017, 6, 271–288. [CrossRef]
36. Almouhanna, A.; Quintero-Araujo, C.L.; Panadero, J.; Juan, A.A.; Khosravi, B.; Ouelhadj, D. The location routing problem using
electric vehicles with constrained distance. Comput. Oper. Res. 2020, 115, 104864. [CrossRef]
37. Jin, F.; Yao, E.; An, K. Analysis of the potential demand for battery electric vehicle sharing: Mode share and spatiotemporal
distribution. J. Transp. Geogr. 2020, 82, 102630. [CrossRef]
38. Chen, Y.R.; Li, D.C.; Zhang, Z.C.; Wahab, M.I.M.; Jiang, Y.S. Solving the battery swap station location-routing problem with a
mixed fleet of electric and conventional vehicles using a heuristic branch-and-price algorithm with an adaptive selection scheme.
Expert Syst. Appl. 2021, 186, 115683. [CrossRef]
39. Yang, Y.; He, K.; Wang, Y.P.; Yuan, Z.Z.; Yin, Y.H.; Guo, M.Z. Identification of dynamic traffic crash risk for cross-area freeways
based on statistical and machine learning methods. Phys. A Stat. Mech. Its Appl. 2022, 595, 127083. [CrossRef]
40. Bayliss, C. Machine learning based simulation optimisation for urban routing problems. Appl. Soft Comput. 2021, 105, 107269.
[CrossRef]
41. Yang, Y.; Yuan, Z.Z.; Chen, J.J.; Guo, M.Z. Assessment of osculating value method based on entropy weight to transportation
energy conservation and emission reduction. Environ. Eng. Manag. J. 2017, 16, 2413–2423. [CrossRef]
42. Yang, Y.; Wang, K.; Yuan, Z.Z.; Liu, D. Predicting freeway traffic crash severity using XGBoost-Bayesian network model with
consideration of features interaction. J. Adv. Transp. 2022, 2022, 4257865. [CrossRef]
Sustainability 2022, 14, 11681 30 of 31

43. Barco, J.; Guerra, A.; Munoz, L.; Quijano, N. Optimal routing and scheduling of charge for electric vehicles: A case study. Math.
Probl. Eng. 2017, 2017, 1–16. [CrossRef]
44. Kyriakakis, N.A.; Stamadianos, T.; Marinaki, M.; Marinakis, Y. The electric vehicle routing problem with drones: An energy
minimization approach for aerial deliveries. Clean. Logist. Supply Chain 2022, 4, 100041. [CrossRef]
45. Zhang, H.; Tang, L.; Yang, C.; Lan, S.L. Locating electric vehicle charging stations with service capacity using the improved whale
optimization algorithm. Adv. Eng. Inform. 2019, 41, 100901. [CrossRef]
46. Yu, V.F.; Jodiawan, P.; Gunawan, A. An Adaptive Large Neighborhood Search for the green mixed fleet vehicle routing problem
with realistic energy consumption and partial recharges. Appl. Soft Comput. 2021, 105, 107251. [CrossRef]
47. Erdem, M. Optimisation of sustainable urban recycling waste collection and routing with heterogeneous electric vehicles. Sustain.
Cities Soc. 2022, 80, 103785. [CrossRef]
48. Breunig, U.; Baldacci, R.; Hartl, R.F.; Vidal, T. The electric two-echelon vehicle routing problem. Comput. Oper. Res. 2019, 103,
198–210. [CrossRef]
49. Chakraborty, N.; Mondal, A.; Mondal, S. Intelligent charge scheduling and eco-routing mechanism for electric vehicles: A
multi-objective heuristic approach. Sustain. Cities Soc. 2021, 69, 102820. [CrossRef]
50. Felipe, Á.; Ortuño, M.T.; Righini, G.; Tirado, G. A heuristic approach for the green vehicle routing problem with multiple
technologies and partial recharges. Transp. Res. Part E Logist. Transp. Rev. 2014, 71, 111–128. [CrossRef]
51. Froger, A.; Mendoza, J.E.; Jabali, O.; Laporte, G. Improved formulations and algorithmic components for the electric vehicle
routing problem with nonlinear charging functions. Comput. Oper. Res. 2019, 104, 256–294. [CrossRef]
52. Jie, W.C.; Yang, J.; Zhang, M.; Huang, Y.X. The two-echelon capacitated electric vehicle routing problem with battery swapping
stations: Formulation and efficient methodology. Eur. J. Oper. Res. 2019, 272, 879–904. [CrossRef]
53. Hof, J.; Schneider, M.; Goeke, D. Solving the battery swap station location-routing problem with capacitated electric vehicles
using an AVNS algorithm for vehicle-routing problems with intermediate stops. Transportation Res. Part B Methodol. 2017, 97,
102–112. [CrossRef]
54. Ma, B.S.; Hu, D.W.; Chen, X.Q.; Wang, Y.; Wu, X. The vehicle routing problem with speed optimization for shared autonomous
electric vehicles service. Comput. Ind. Eng. 2021, 161, 102214. [CrossRef]
55. Pelletier, S.; Jabali, O.; Laporte, G. The electric vehicle routing problem with energy consumption uncertainty. Transp. Res. Part B
Methodol. 2019, 126, 225–255. [CrossRef]
56. Zhu, Z.H.; Gao, Z.Y.; Zheng, J.F.; Du, H.M. Charging station location problem of plug-in electric vehicles. J. Transp. Geogr. 2016,
52, 11–22. [CrossRef]
57. Keskin, M.; Çatay, B. Partial recharge strategies for the electric vehicle routing problem with time windows. Transp. Res. Part C
Emerg. Technol. 2016, 65, 111–127. [CrossRef]
58. Cataldo-Diaz, C.; Linfati, R.; Escobar, J.W. Mathematical model for the electric vehicle routing problem considering the state of
charge of the batteries. Sustainability 2022, 3, 1645. [CrossRef]
59. Wang, Y.; Li, Q.; Guan, X.Y.; Fan, J.X.; Liu, Y.; Wang, H.Z. Collaboration and resource sharing in the multidepot multiperiod
vehicle routing problem with pickups and deliveries. Sustainability 2020, 12, 5966. [CrossRef]
60. Li, J.J.; Fang, Y.H.Q.; Tang, N. A cluster-based optimization framework for vehicle routing problem with workload balance.
Comput. Ind. Eng. 2022, 169, 108221. [CrossRef]
61. Chen, Q.; Pan, X.Y.; Liu, F.; Xiong, Y.; Li, Z.T.; Tang, J.J. Reposition optimization in free-floating bike-sharing system: A case study
in Shenzhen City. Phys. A Stat. Mech. Its Appl. 2022, 593, 126925. [CrossRef]
62. Tang, J.J.; Hu, J.; Hao, W.; Chen, X.Q.; Qi, Y. Markov Chains based route travel time estimation considering link spatio-temporal
correlation. Phys. A Stat. Mech. Its Appl. 2020, 545, 123759. [CrossRef]
63. Liao, W.Z.; Zhang, L.Y.; Wei, Z.Z. Multi-objective green meal delivery routing problem based on a two-stage solution strategy. J.
Clean. Prod. 2020, 258, 120627. [CrossRef]
64. Eydi, A.; Ghasemi-Nezhad, S.A. A bi-objective vehicle routing problem with time windows and multiple demands. Ain Shams
Eng. J. 2021, 12, 2617–2630. [CrossRef]
65. Srivastava, G.; Singh, A.; Mallipeddi, R. NSGA-II with objective-specific variation operators for multiobjective vehicle routing
problem with time windows. Exp. Syst. Appl. 2021, 176, 114779. [CrossRef]
66. Katoch, S.; Chauhan, S.S.; Kumar, V. A review on genetic algorithm: Past, present, and future. Multimed. Tools Appl. 2021, 5,
8091–8126. [CrossRef]
67. Wang, Y.; Zhang, S.L.; Assogba, K.; Fan, J.X.; Xu, M.Z.; Wang, Y.H. Economic and environmental evaluations in the two-echelon
collaborative multiple centers vehicle routing optimization. J. Clean. Prod. 2018, 197, 443–461. [CrossRef]
68. Li, Q.; Cao, Z.H.; Ding, W.P.; Li, Q. A multi-objective adaptive evolutionary algorithm to extract communities in networks. Swarm
Evol. Comput. 2020, 52, 100629. [CrossRef]
69. Khoo, T.S.; Mohammad, B.B. The parallelization of a two-phase distributed hybrid ruin-and-recreate genetic algorithm for solving
multi-objective vehicle routing problem with time windows. Expert Syst. Appl. 2021, 168, 14408. [CrossRef]
70. Koohestani, B. A crossover operator for improving the efficiency of permutation-based genetic algorithms. Expert Syst. Appl.
2020, 151, 113381. [CrossRef]
71. Wang, Y.; Wang, X.W.; Guan, X.Y.; Tang, J.J. Multidepot recycling vehicle routing problem with resource sharing and time window
assignment. J. Adv. Transp. 2021, 2021, 2327504. [CrossRef]
Sustainability 2022, 14, 11681 31 of 31

72. Giallanza, A.; Puma, G.L. Fuzzy green vehicle routing problem for designing a three echelons supply chain. J. Clean. Prod. 2020,
259, 120774. [CrossRef]
73. Martínez-Puras, A.; Pacheco, J. MOAMP-Tabu search and NSGA-II for a real bi-objective scheduling-routing problem. Knowl.
Based Syst. 2016, 112, 92–104. [CrossRef]
74. Liu, Y.; Zhu, N.B.; Li, K.L.; Li, M.Q.; Zheng, J.H.; Li, K.Q. An angle dominance criterion for evolutionary many-objective
optimization. Inf. Sci. 2020, 509, 376–399. [CrossRef]
75. Zhang, M.Q.; Wang, L.; Guo, W.A.; Li, W.Z.; Li, D.Y.; Hu, B.; Wu, Q.D. Many-objective evolutionary algorithm based on relative
non-dominance matrix. Inf. Sci. 2021, 547, 963–983. [CrossRef]
76. Ghannadpour, S.F.; Zandiyeh, F. A new game-theoretical multi-objective evolutionary approach for cash-in-transit vehicle routing
problem with time windows (A Real life Case). Appl. Soft Comput. 2020, 93, 106378. [CrossRef]
77. Maiyar, L.M.; Thakkar, J.J. Environmentally conscious logistics planning for food grain industry considering wastages employing
multi objective hybrid particle swarm optimization. Transp. Res. Part E Logist. Transp. Rev. 2019, 127, 220–248. [CrossRef]
78. Li, Y.B.; Soleimani, H.; Zohal, M. An improved ant colony optimization algorithm for the multi-depot green vehicle routing
problem with multiple objectives. J. Clean. Prod. 2019, 227, 1161–1172. [CrossRef]

You might also like