An adaptive artificial bee colony algorithm enhanced by Deep Q-Learning for milk-run vehicle scheduling problem based on supply hub

Abstract

To meet market challenges and reduce costs, supply hubs are broadly applied to achieve just-in-time production. In this paper, a milk-run vehicle scheduling problem considering the ratio of parts is proposed based on the supply hub mode. A mathematical model is therefore established, aiming to minimize the energy consumption of electric vehicles and the time penalty of early or late distribution simultaneously in the scheduling horizon. To solve the proposed problem which is inherently NP-hard, an adaptive artificial bee colony algorithm enhanced by deep Q-learning network (AABC-DQN) is constructed, where the deep Q-learning network is used to reasonably select the neighborhood search operators to systematically expand the search range. Experimental results have demonstrated the superiority of the proposed AABC-DQN, which outperforms the benchmark algorithms in both solution quality and convergence ability.