(189ad) A New Proactive Methodology for Robust Berth Planning of Container Vessels

The rise in international trade is significantly observed over the years and one of the major transport mode used is the container ships, which also have potential impact on chemical industry. With many ships arriving at the port periodically, the operation schedule at container ports is becoming tighter. Over the past several years, a reduction in the productivity of port has been observed due to inefficient usage of the available resources. Amongst all the operations, the prime most important is an optimal berth allocation of the ship and yard allocation for the containers. The berth location of ships and the yard location of containers are inter-dependent on each other based on the transshipment of the containers. These operations are based on constraints like allowable time window for berthing, quay crane capacity and allowable process time for the ships. Thus, there is a mighty importance of scheduling of port operations and optimizing the available resources at the port terminal. Mathematical optimization has been used to improve the operational capability of port terminals using various realistic constraints.

Conventional berth allocation and planning models have difficulties to handle uncertainty of vessel arrival while constructing a nominal berth plan or nominal berth allocation. However, in reality, timetables become more and more vulnerable to disturbances. The delay of a single vessel might interrupt the entire schedule and make it very difficult or even impossible to recover. To deal with disturbances in transportation schedules, two approaches are getting more and more attention: (1) proactive robustness, which incorporates buffer times into strategic or tactical timetables to deal with disturbances and thus to prevent delay propagation through a well set schedule, and (2) reactive disruption management, which is concerned with operational recovery after a disruption. The research in this paper focuses on incorporating proactive robustness into the nominal berth planning.

Container vessels might arrive earlier or later than their nominal arrival time due to all kinds of uncertainties during port operations. Therefore, the customers and terminal operators agree upon an arrival window, which is placed around the nominal arrival time. The time interval between the nominal arrival and departure time is the nominal vessel process time. Existing publications are either focused on planning problem without considering realistic flexible arrival and departure time windows [1, 2], or using fixed maximal deviation from nominal arrival time that is still within the arrival window of vessels [3].

The problem is hence to develop a time window-based berth allocation planning model that minimizes the maximally required operation time for all scenarios where vessels arrive within their arrival time windows and depart within their departure time windows. The flexible arrival and departure time windows have been into account specifically as control variables. A robust mixed integer nonlinear programming (MINLP) model is developed, which explicitly incorporates the operation time agreements and minimizes the maximally required operation time. The general objective is to minimize the maximally required operation time while satisfy all the transportation demand and operation constrains. Commercial solvers such as BARON, DICOPT and ANTIGONE have been employed and compared to obtain the optimal solution of the developed MINLP planning model. Computational results of the case study demonstrate the efficacy of the developed planning model, and a significant reduction in the maximal operation time can be obtained.

References

1. Wang, F. and A. Lim, A stochastic beam search for the berth allocation problem. Decis. Support Syst., 2007. 42(4): p. 2186-2196.

2. Cordeau, J.-F., et al., Models and Tabu Search Heuristics for the Berth-Allocation Problem. Transportation Science, 2005. 39(4): p. 526-538.

3. Hendriks, M., et al., Robust cyclic berth planning of container vessels. OR Spectrum, 2010. 32(3): p. 501-517.