3RD INTERNATIONAL CONGRESS ON TECHNOLOGY - ENGINEERING & SCIENCE - Kuala Lumpur - Malaysia (2017-02-09)

Fleet Optimization Of Electric Scooter Sharing Systems Via Simulation
Electric Scooter Sharing Systems (ESSS) are environment-friendly transportation alternatives for short trips that are often inadequately served by fixed route public transit systems. ESSS do not only solve the “last mile” problem of the public transportation systems but they also offer extra benefits to existing bicycle sharing customers. The basic premise of the ESSS is allowing people to rent an e-scooter at one of the automatic rental stations scattered around the city for a short journey and return them at any station in the city. The literature existing on similar shared use concepts is for bike-sharing and car-sharing. These are public vehicle sharing systems that were first implemented. The study focused on determining service level requirements at each bike sharing station and the design of optimal vehicle routes to rebalance the inventory, see [1]. Other work studied strategic problems, such as [2] and [3] who addressed the question of bike rental stations’ capacity and locations. In [4], a simulation approach was used to extend bike sharing research domain. In [5], a fleet size and composition optimization model with cost constraints was developed for the car-sharing system. This study investigates the optimal initial vehicle deployment for a fully automated ESSS implemented in a city with the highest density of universities in Taiwan. With limited empirical demand and usage information, we simulated the operation of the system, with special emphasis on the distribution of the critical demand parameters (trip rates, trip lengths and trip durations) and supply parameters (number of e-scooter and charging dock and recharge protocol), to determine the optimal number of e-scooters under different potential demand situations. In this study, we also present a cost constrained e-scooter sharing system design that can maintain the high level of system availability, and help the ESSS service providers gain important insights before full field deployment. By adjusting input parameters, numerous scenarios were simulated for sensitivity analysis. Based on the simulation results, we found that the optimal initial number of e-scooters required by the system was significantly sensitive to trip rate, trip length, and activity duration. The factors of trip rate and activity duration exercise a positive effect on the number of e-scooters. Increasing trip rate by triple has the equivalent effect on e-scooter requirements. An application of this model could be to design the size of a station based on demand variables that one might observe. Optimal initial deployment of e-scooter for users is crucial to the system, balancing system performance, capital cost, and user convenience. The success factor of an electric scooter sharing system is its ability to meet the fluctuating demand for e-scooter at the station. E-scooter availability problems are important to investigate because ultimately they will affect customer satisfaction and confidence to use the services of ESSS.
Silvia Merdikawati, Shi-Woei Lin, Amalia Suzianti