Combined Berth Allocation and Yard Planning


Jan Tijmen Udding
  • Speaker
Department of Mechanical Engineering, Eindhoven University of Technology

Event Information

Type
Research Seminar
Programme
Logistics
Date
Wed. 19 Nov. 2008
Contact
Time
12:00-13:00 hours
E-mail
Location
Mandeville Building T10-67
Number


Abstract

Container vessels typically call at a container terminal in a cyclic rotation. The average call size, i.e. the number (and types) of containers to be loaded and unloaded is known. Containers to be loaded onto such a vessel arrive at the terminal between a few weeks up to a few hours prior to its departure time. In a straddle carrier terminal, containers loaded onto a particular ship are stacked together in designated areas.  Preferably, given the locations of the containers on the yard, vessels should berth close to the containers to be loaded in order to reduce the driving distance of the straddle carriers. We have captured this problem as an MILP that finds a solution in just a few seconds, which means that it can be used at an operational level.

 
Some designated areas are fixed, for example those for reefer containers. Areas for the general purpose containers can, however, freely be chosen. Hence, actually the problem is two-sided: in addition to the problem above, one would like to stack containers close to the berthing location of a calling vessel as well. In this problem, both the designated areas in the yard and the berthing locations of the vessels are decision variables. Although we cannot phrase this as a Linear Programming Problem we show how we can phrase it as two independent problems, one an MILP and one an LP, that can be solved iteratively. Since the performance of this alternating procedure turns out to depend heavily on the initial condition, an additional MILP is developed. This MILP allocates fixed groups of containers, rather than variably-sized groups, and appears to generate a proper initial condition for the alternating procedure.
 
Contact information:
Prof.dr. M.B.M. de Koster
Email
René de Koster
Professor of Logistics and Operations Management
  • Coordinator