Location optimization for multiple types of charging stations for electric scooters

doi:10.1016/j.asoc.2018.02.038

Applied Soft Computing

Volume 67, June 2018, Pages 519-528

https://doi.org/10.1016/j.asoc.2018.02.038 Get rights and content

Highlights

•
A Multi-Objective Particle Swarm Optimization for location-allocation problem.
•
The characteristics consider the population density and land cost.
•
The compound site selection considering recharging station and battery exchange station for e-scooter.

Abstract

The difference between traditional scooters and electric scooters is the convenience of refueling and charging process. Designing a complete infrastructure system is a necessary step if efforts to promote e-scooters are to meet with success. This study discusses the optimal location problem of locating charging stations—which is generally considered a location-allocation problem. There are two types of charging stations: charge stations and battery-exchange stations. The only one model in determining the location of either type of station tends to decrease the traditional utility compared to compound model traditional as this method tends not to set stations where they would serve the greatest number of customers. In addition, population density and land cost should be taken into account in determining where stations are set. We, therefore, propose a method that accounts for differences in population density and land cost in order to solve a multi-objective problem with maximum utility at minimum cost. A mathematical model is developed in which constraints pertaining to capacity and distance are considered. To find an optimal parameter for Multi-Objective Particle Swarm Optimization (MOPSO), generational distance (GD), maximum spread, spacing, and diversity metrics are applied. Finally, we research an angle-based focus method and determine the extent to which stations would be used in order to determine the optimal proportions of charging stations and battery-exchange stations. Moreover, according to the analysis we found that the installed ratio model of BES/BCS (Battery-exchange stations/Battery charging stations) is 6:5 in the downtown area and 1:6 in the outskirts of high population density areas. Besides, the BES/BCS ratio model is 1:13 in the downtown and only BCSs are installed in the outskirts of the low population density area.

Graphical abstract

A Multi-Objective Particle Swarm Optimization for location-allocation problem is developed to minimize total cost and maximize service capacity considering land price, service distance, and installation capacity at each station.

Introduction

About 15 percent of manmade carbon dioxide comes from cars, trucks, airplanes, ships, and other vehicles. Reducing transportation emissions, therefore, is one of the most critical steps in fighting global warming. Environmentally friendly means of personal transportation such as electric cars, e-scooters, and electric bikes are among the solutions available and have become the subject of increasing attention as global warming has worsened. In particular, e-scooters may constitute a viable form of transportation given that they create less pollution than fuel scooters do [1]. However, in order to promote e-cars, e-scooters, and e-bikes effectively, it is necessary to consider not only the performance and efficiency gains associated with the power trains used in these vehicles but also the customer’s experience.

There are two issues in regard to determining the location of BES/BCS systems: station type and station location. According to Wang and Lin [2], the main purpose of facility planning is that of determining the locations of multi-type stations, i.e., those that combine BES and BCS systems to that customers can access various stations and convenient battery refills. You and Hsieh [3] identified number, location, and type as three key parameters for stations that are necessary to consider in planning BES/BCS systems. They also concluded that location and station type should be given the most weight when stringent cost limitations are involved. Two cost limitations are discussed in the present study: the land cost of the location and the installation cost of battery-refill systems. Undoubtedly, the land cost is a fundamental issue to address in the location-allocation problem. Doong, Lai, and Wu stated that final location decisions are affected by various land costs associated with candidate areas. It should be expected, therefore, that land cost together with the installation costs of the two battery-refill systems, BES and BCS, are matters of great concern in location-allocation decisions. When only one type of battery-refill system is used at a station, however, the extent to which that station used can be expected to be limited.

In the present investigation, our goal is to understand the optimal facility location decision in regard to the highest numbers of times a station is used of the refill system. A hybrid refill system allocation in various land cost conditions can provide an optimal solution to the problem presented by the short battery life of e-scooters and the inconvenience of refilling, thereby providing a foundation for increasing the sale and use of this form of transportation.

This paper only focuses on the locations and allocations of battery-exchange station (BES) systems and battery-charging station (BCS) systems for e-scooters. Although charging and battery-exchange stations could provide services to almost all types of electric mobility devices, the assumptions and problems should be studied separately. The objective is to maximize the cover rate of electric scooters’ customers instead of other types of electric mobility devices such as vehicles and bicycles, because maximum distance acceptable to customer for various mobility devices are different. For example, electric vehicles may have longer acceptable distance than electric scooters. Furthermore, the socket outlet, and required voltage are substantial different in electric mobility devices. The required power loading for the charging and battery-exchange stations is relatively small comparing to general electric cars. Charging electric scooters requires a regular household outlet of 115VAC, 15A, and produces about 1.5 kW, and the charge time is 7–30 h depending on battery size. Usually electric scooters prefer to have the battery-exchange service or overnight charging requirements. On other hand, electric vehicles require 230VAC, 30A two pole, and produces about 7 kW to charge a mid-sized electric vehicle in 4–5 hours. In this case, candidate locations such as convenient store, supermarket, and malls have the capabilities of setup the extra charging services using regular household outlet.

In the present study, we review the literature on the location-allocation problem, present a Multi-Objective Particle Swarm Optimization (MOPSO) algorithm, describe the design and verify our mathematical model and algorithm, and analyze the critical parameters and Pareto-optimal front. We also provide a conclusion and suggest directions for future research.

Section snippets

Related work

Cooper defined the location-allocation problem in terms of the need to provide goods or services to satisfy a spatially dispersed demand. Based on this definition, in regard to the subject of the present study, when there is a demand for battery-refill services at a large number of widely dispersed sites, it is usually impossible to provide the service everywhere. Therefore, it is necessary to engage in a process of decision-making aimed at effecting optimization in a continual way under cost

Mathematical model

The problem considered in the present study can be described as follows. There are i customer nodes and j charging stations, and each of the latter is located in either a high- or a low-price area. The objective is to minimize the total cost of setting charging stations and to maximize the total number of customers served.

I	Customer nodes	$i = {1,2, \dots, \| I \|} \in I$
J	Charging stations	$j = {1,2, \dots, \| J \|} \in J$
A	Area	$a = {1,2 \dots, \| A \|} \in A$

Parameters:

$d_{i j}^{a}$	Distance from customer i to charging station j in area a
$P_{a}$	Land cost of area a

Model evaluation

In this section, we describe how we implemented the mathematical models and verified the algorithm coding. The generational distance (GD), maximum spread (MS), spacing (S), and diversity metric (Δ) were used to determine the optimal learning factor with the MOPSO algorithm. The optimal learning factor was inputted into the algorithm to explore the generated Pareto front by using a dual-objective PSO. The front solution was analyzed in three stages: stage 1, which involved a case analysis of

Conclusion and future research

In this study, a multi-objective MILP model for the location-allocation optimization problem by applying the PSO algorithm was proposed. The location and allocation decisions are optimized in reference to minimum total cost and maximum service capacity as two objectives and in reference to land price, service distance, and installation capacity at each station as the limitations. This study used Ilog Cplex 9.0 to develop the mathematical optimization model as a comparison to the PSO algorithm.

References (22)

Y.-W. Wang et al.
Locating multiple types of recharging stations for battery-powered electric vehicle transport
Transp. Res. Part E: Logist. Transp. Rev.
(2013)
P.-S. You et al.
A hybrid heuristic approach to the problem of the location of vehicle charging stations
Comput. Ind. Eng.
(2014)
R.L. Church et al.
Reserve selection as a maximal covering location problem
Biol. Conserv.
(1996)
C. Upchurch et al.
Comparing the p-median and flow-refueling models for locating alternative-fuel stations
J. Transp. Geogr.
(2010)
M. Kuby et al.
The flow-refueling location problem for alternative-fuel vehicles
Socioecon. Plann. Sci.
(2005)
Y.-W. Wang et al.
Locating road-vehicle refueling stations
Transp. Res. Part E: Logist. Transp. Rev.
(2009)
H. Gavranović et al.
Optimizing the electric charge station network of EŞARJ
Procedia Comput. Sci.
(2014)
J. Gang et al.
A multi-objective bi-level location planning problem for stone industrial parks
Comput. Oper. Res.
(2015)
A. Zeinalzadeh et al.
Optimal multi objective placement and sizing of multiple DGs and shunt capacitor banks simultaneously considering load uncertainty via MOPSO approach
Int. J. Electr. Power Energy Syst.
(2015)
N. Adnan et al.
An overview of electric vehicle technology: a vision towards sustainable
Green Mark. Environ. Responsibility Mod. Corp.
(2017)

S. Hakimi

Optimum distribution of switching centers in a communication network and some related graph theoretic problems

Oper. Res.

(1965)

Cited by (62)

Supplier selection of companies providing micro mobility service under type-2 neutrosophic number based decision making model
2024, Expert Systems with Applications
Environmental approaches play an active role in the production platform as well as in the supply chain. Studies that deal with supplier selection problems in the supply chain with a green focus have recently intensified. In this research, the supplier selection problem of companies providing “last mile” service, which is seen as the last step of transportation, is handled with a green focus. The supplier selection problem of these companies is differentiated according to strategic and non-strategic product groups. In this study, a green-oriented supplier selection proposal is presented for both the strategic product group and the non-strategic product group. The supplier selection problem is accepted as a multi criteria decision making problem (MCDM). Thus, the type-2 neutrosophic numbers (T2NNs) - based CRiteria Importance Through Intercriteria Correlation (CRITIC) – based Multi-Attributive Border Approximation Area Comparison (MABAC) hybrid model is applied to evaluate the green-oriented supplier. The motivation of the study is to determine the best alternative supplier type for strategic and non-strategic products groups. In this context, two cases are created, ten criteria are selected, and three supplier types are determined. Supplier types are as follows: (i) Classical suppliers, (ii) Hybrid-Classical to green centric suppliers, and (iii) Start-Ups & green centric suppliers. To devise a solution for addressing the research problem, the selection of the optimal green supplier type was accomplished through a recommended T2NN-CRITIC-MABAC hybrid model. Accordingly, for both strategic and non-strategic product groups, micro-mobility service providers identified Hybrid-Classical to Green Centric Suppliers as the most suitable green supplier type. In addition, the findings are strengthened by creating twenty-two sensitivity analysis scenarios. In conclusion, the T2NN-CRITIC-MABAC hybrid method introduced in this study represents a methodologically diverse, fuzzy logic-based approach with practical implications for supplier selection in the micro mobility services sector. Its application and empirical testing contribute to the literature and provide valuable insights for decision-makers in the industry.
An e-scooter service region and fleet allocation design problem with elastic demand
2024, Transportation Research Part D: Transport and Environment
Despite the rapid growth of micromobility, and e-scooters in particular, there is no systematic approach for designing a contiguous service region with fleet allocation and responsive demand. We propose a non-rooted maximum weighted connected subgraph model with demand response to deal with this need, the first such model that does not require enumerating cycles or node pairs. The nonlinear demand response function is piecewise-linearized to obtain a mixed integer linear program and solved using an open-source solver, which is computationally efficient even for a Manhattan-sized study area with run times on the order of minutes for 318 zones. Two sets of experiments are conducted, the first on a 4-by-4 grid region to verify the model, and a second on the Manhattan region using realistic data. The results show that (1) there can exist thresholds where a strategy may switch from zone expansion to fleet expansion; and (2) reveals areas in Manhattan that should be avoided.
A multi-objective model to design shared e-kick scooters parking spaces in large urban areas
2024, Journal of Transport Geography
In recent years, the micromobility and the usage of shared electric kick scooters (e-kscooters) have been constantly growing, especially for systematic and recreational trips in large urban areas. Micromobility might be seen as a well-suited last-mile solution by providing a flexible travel service connection with public transport and MaaS (Mobility as a Service), in general. However, there is a need for implementing adequate regulations regarding safety aspects and shared e-kscooter parking locations, but also for meeting the user requirements. The choice of optimal shared e-kscooter parking locations could help decision-makers to regulate unmanaged dock-less shared e-kscooter parking spots that could generate issues for other road users. To this end, in this paper, a novel multi-objective Micromobility Maximal Coverage Parking Location model (M-MCPL) is developed. The model has been solved by applying an elitist Genetic Algorithm that returns the optimal shared e-kscooter parking locations based on the following objective functions: i) the maximization of the population coverage; ii) the maximization of multimodal accessibility coverage (i.e., bus, railway, and metro modes); iii) the maximization of the attraction coverage considering the most relevant points of interest for each corresponding zone in large urban areas. The proposed M-MCPL model has been applied to the case of Rome (Italy) and results suggest priorities for the shared e-kscooter parking locations design. Furthermore, the proposed model is flexible and can be considered as a decision support tool for decision-makers when planning dedicated services in different large urban areas. For that purpose, we conducted the sensitivity analysis by focusing on the single-objective model in which decision-makers might be interested in providing only high accessibility to transport services or maximizing potential demand.
Identifying barriers and benefits of shared e-scooters in promoting sustainability in Kuwait using Delphi and Force Field Analysis Methods
2024, Journal of Engineering Research (Kuwait)
People are wasting time during their way to their destination due to traffic jam. Shared electronic scooter is ana alternative transport mode that saves time. This mode of transportation widely spread after COVID-19 pandemic. This research investigates the main opportunities and barriers of adopting this technology in Kuwait from three sides which are social, environmental, economic, and safety aspects taking into consideration the travel behavior and urban infrastructure sides. Experts’ opinions were collected from two rounds of the Delphi questionnaire. The collected data was analyzed using Force Field Analysis Method in order to get a precises prediction of the most significant facilitators and barriers after the period of COVID-19 to figure out the importance of the sustainability of the e-scooters. According to the urban infrastructure side, the e-scooters adoption was favorable from all aspects except the economic aspect. On the other hand, travel behavior was challenging, it was unfavorable form environmental and safety aspects. It’s recommended to regulate more laws that ensure all road users’ safety.
Sparse trip demand prediction for shared E-scooter using spatio-temporal graph neural networks
2023, Transportation Research Part D: Transport and Environment
The shared electric scooter (E-scooter) is an emerging micro-mobility mode in sustainable cities. Accurate hourly trip demand prediction is critical for effective service maintenance, but it poses a challenge due to the dynamic distribution influenced by urban complexity. We propose a model, the Sparse Diffusion Convolutional Gated Recurrent Unit (SpDCGRU), which incorporates diffusion convolution layers into the gated recurrent unit (GRU) model, enabling the simultaneous capture of spatio-temporal dependencies. Tackling the data in Louisville, Kentucky, USA, we demonstrate that spatial data reclustering and fusion loss training strategies contribute to the prediction performance. Moreover, the periodic and weather features positively impact predicting the low and high trip demand levels, respectively. Our model outperforms others in terms of overall performance and each trip demand level, with a 4.75% improvement in the mean absolute error (MAE) compared to the graph convolutional recurrent network (GCRN).
Visualization and bibliometric analysis of e-bike studies: A systematic literature review (1976–2023)
2023, Transportation Research Part D: Transport and Environment
Electric bicycles (also named e-bikes or e-scooters) are considered as a key component of the replacement of low-carbon transportation. To gain knowledge of current situation and anticipated shifts in the field of e-bikes timely and comprehensively, firstly, 2,101 articles related to e-bikes from the WoS-Core-based repository are extracted. Then, mapping knowledge domain tools are performed to conduct bibliometric analysis of selected articles. The results show that energy models and charging system, behaviour and safety, environmental and health, usage and influencing are the major concerns through three development process, namely primary development-based, stable development-based, and rapid development-based stage. Finally, research gaps, future studies and discussion on the concerns of e-bikes are presented.

View all citing articles on Scopus

View full text

Location optimization for multiple types of charging stations for electric scooters

Highlights

Abstract

Graphical abstract

Introduction

Section snippets

Related work

Mathematical model

Model evaluation

Conclusion and future research

Transp. Res. Part E: Logist. Transp. Rev.

Comput. Ind. Eng.

Biol. Conserv.

J. Transp. Geogr.

Socioecon. Plann. Sci.

Transp. Res. Part E: Logist. Transp. Rev.

Procedia Comput. Sci.

Comput. Oper. Res.

Int. J. Electr. Power Energy Syst.

An overview of electric vehicle technology: a vision towards sustainable

Green Mark. Environ. Responsibility Mod. Corp.

Optimum distribution of switching centers in a communication network and some related graph theoretic problems

Oper. Res.