Title: Assignment strategy selection for multi-car elevator group control using reinforcement learning
Authors: Taichi Uraji; Kenichi Takahashi
Addresses: Graduate school of Information Science, Hiroshima City University, 3-4-1, Ozukahigashi, Asaminami-ku, Hiroshima, 731-3194, Japan. ' Graduate school of Information Science, Hiroshima City University, 3-4-1, Ozukahigashi, Asaminami-ku, Hiroshima, 731-3194, Japan
Abstract: This paper discusses the group control of elevators in the web monitoring system for improving efficiency and saving energy; an efficient control method for multi-car elevator using reinforcement learning is proposed. In the method, the control agent selects the best strategy among three strategies, namely distance-strategy, passenger-strategy, and zone-strategy, according to traffic flow. The control agent takes the number of total passengers and the distance from the departure floor to the destination floor of a call into account. Through experiments, the performance of the proposed method is shown; the average service time of the proposed method is compared with the average service time for the cases where the car assignment is made by each of the three strategies.
Keywords: multi-car elevator systems; multi-car lift systems; reinforcement learning; group control; elevators; lifts; web monitoring; traffic flow; control agents; average service time; assignment strategy; elevator control; lift control.
International Journal of Knowledge and Web Intelligence, 2012 Vol.3 No.2, pp.163 - 179
Published online: 04 Sep 2014 *Full-text access for editors Access for subscribers Purchase this article Comment on this article