Dynamic Scheduling of Handling Equipment at Automated Container Terminals

In this paper we consider the problem of integrated scheduling of various types of handling equipment at an automated container terminal in a dynamic environment. This means that the handling times are not known exactly beforehand and that the order in which the different pieces of equipment handle the containers need not be specified completely in advance. Instead, (partial) schedules may be updated when new information on realizations of handling times becomes available. We present an optimization based Beam Search heuristic and several dispatching rules. An extensive computational study is carried out to investigate the performance of these solution methods under different scenarios. The main conclusion is that, in our tests, the Beam Search heuristic performs best on average, but that some of the relatively simple dispatching rules perform almost as good. Furthermore, our study indicates that it is effective important to base a planning on a long horizon with inaccurate data, than to update the planning often in order to take newly available information into account.