Modeling and Optimization of Multiaction Dynamic Dispatching Problem for Shared Autonomous Electric Vehicles

Abstract

The fusion of electricity, automation, and sharing is forming a new Autonomous Mobility-on-Demand (AMoD) system in current urban transportation, in which the Shared Autonomous Electric Vehicles (SAEVs) are a fleet to execute delivery, parking, recharging, and repositioning tasks automatically. To model the decision-making process of AMoD system and optimize multiaction dynamic dispatching of SAEVs over a long horizon, the dispatching problem of SAEVs is modeled according to Markov Decision Process (MDP) at first. Then two optimization models from short-sighted view and farsighted view based on combinatorial optimization theory are built, respectively. The former focuses on the instant and single-step reward, while the latter aims at the accumulative and multistep return. After that, the Kuhn–Munkres algorithm is set as the baseline method to solve the first model to achieve optimal multiaction allocation instructions for SAEVs, and the combination of deep Q-learning algorithm and Kuhn–Munkres algorithm is designed to solve the second model to realize the global optimization. Finally, a toy example, a macrosimulation of 1 month, and a microsimulation of 6 hours based on actual historical operation data are conducted. Results show that (1) the Kuhn–Munkres algorithm ensures the computational effectiveness in the large-scale real-time application of the AMoD system; (2) the second optimization model considering long-term return can decrease average user waiting time and achieve a 2.78% increase in total revenue compared with the first model; (3) and integrating combinatorial optimization theory with reinforcement learning theory is a perfect package for solving the multiaction dynamic dispatching problem of SAEVs.

Multi-task dispatch of shared autonomous electric vehicles for Mobility-on-Demand services – combination of deep reinforcement learning and combinatorial optimization method

“…Although our previous research Wang and Guo (2021) is the first to model these three sub-problems in one framework by the reinforcement learning method, this study differs in 3 aspects: (1) Although the fleet dispatching process is also modeled by the MDP framework, the reward function of delivery action in this paper is improved by considering both order revenue and user satisfaction; (2) Three sub-problems are transferred into a 3-stage decision process rather than a 1-stage decision process, in which recharging and delivery task assignment are transferred into two individual bipartite graph matching problems separately, and repositioning task assignment is transferred into a max flow problem; (3) When designing the solving algorithm, Edmond-Karp algorithm is first adopted except for Kuhn-Munkres algorithm and deep Q-learning algorithm.…”

“…If vertex and vertex meet the condition , a set called S is defined and made up of directed lines from i to j . If a match is the complete matching of the set S, this match is also the maximum complete matching of the graph G ( Wang and Guo, 2021 ).…”

A survey on emergent trends in the optimization of car‐sharing systems

“…For such a decision problem, an MINLP model is formulated and a genetic algorithm is developed, tested on instances taken from the literature. Wang and Guo (2021) model three decision problems concerning the vehicle assignment, the charging, and the parking assignment, and the VR in a FF sharing system with AVs, maximizing the total revenue. In particular, they model the EAVs dispatching problem through a Markov decision process.…”