Parallel computing in logistics

✍ Scribed by Gilbert Laporte; Roberto Musmanno

• Publisher: Elsevier Science
• Year: 2003
• Tongue: English
• Weight: 48 KB
• Volume: 29
• Category: Article
• ISSN: 0167-8191
• DOI: 10.1016/s0167-8191(03)00041-3

✦ Synopsis

Parallel computing is fastly gaining in importance for the solution of logistics problems. The need to resort to parallel computing is due in part to the large scale of these problems. In addition, a number of real-time logistics applications arising, for example, in the delivery of emergency services and in courier or dial-a-ride services require the computation of solutions in very short computing times, which can be achieved through parallel computing.

The seven articles included in this special issue of Parallel Computing address large scale combinatorial optimization problems encountered in location, vehicle routing and operations planning. The first article, by Campbell, Stiehr, Ernst and Krishnamoorthy describes an enumeration algorithm for hub location problems arising both in passenger and freight transportation. The second contribution, by Garc ı ıa-L o opez, Meli a an-Batista, Moreno-P e erez and Moreno-Vega presents a parallelized scatter search heuristic for the classical p-median problem. It compares three different parallelization strategies. The article of Gendron, Potvin and Soriano describes a parallel hybrid heuristic for the multicommodity capacitated location problem with balancing requirement, a problem central to the planning of container movements. Two procedures based on variable neighbourhood descent and slope scaling are parallelized. The fourth contribution, by Ralphs reports the results of an exact branch-and-cut algorithm for the classical vehicle routing problem. The solver is implemented within SYMPHONY, a library suited for parallel implementation. The next paper, by Caricato, Ghiani, Grieco and E. Guerriero reports on the parallel implementation of two sequential heuristics and a parallel tabu search method for the solution of a constrained pickup and delivery problem. The article of Bortfeldt, Gehring and Mack describes a tabu search algorithm for the solution of a container loading problem. The efficiency of the algorithm is demonstrated on a set of benchmark instances. Finally, the contribution of F. Guerriero and Mancini describes a parallel implementation of the rollout algorithm for the solution of a family of sequential ordering problems arising, namely, in job scheduling.

We thank all authors and referees for their valuable contributions and we hope this special issue will stimulate further research in the development of parallel computing methods, in the fields of transportation and logistics.