Managing ship lock congestion in an inland waterway: A bottleneck model with a service time window

This paper proposes a bottleneck model for managing the ship lock congestion in an inland waterway caused by the uneven temporal distribution of ship traffic demand. Due to safety checks and regular maintenance, the lock only works within a specified time window during a day. A bottleneck model with a service time window is presented to describe the dynamic scheduling process of the ships passing through the lock. The properties of the proposed model are investigated. To alleviate ships' delays at the lock area, two congestion tolling schemes (i.e., first-best and second-best) and a novel administrative scheme of moving the service time window (MST) are explored together with a combined scheme of tolling and MST. The results show that the ship's preferred departure time and the value of time for schedule delay matter in the choice of the counter-measures against ship traffic congestion. MST alone can be an effective substitute for the tolling schemes in most cases and it even outperforms the first-best tolling under certain conditions. This finding is appealing given the academic debate and public controversy raised by congestion tolling. Besides, combining MST with either tolling scheme can always yield higher efficiency. However, the benefits of MST can be marginal or even zero when the difference between the ship's preferred departure time and lock's initial service start time is around a certain value. In such cases, the authority should be cautious about combining any tolling scheme with MST since the implementation costs might outweigh the additional benefits brought by MST.