A MILP-based column generation strategy for managing large-scale maritime distribution problems

Abstract

This paper presents a novel column generation algorithm for managing the logistics activities performed by a fleet of multi-parcel chemical tankers. In our procedure, for providing elementary routes, the conventional dynamic programming routes-generator is replaced by an efficient continuous-time MILP-slave problem. The performance of the decomposition method is evaluated by solving several examples dealing with the operations of a shipping company operating in the Asia Pacific Region. Computational results show that the proposed approach outperforms a pure exact optimization model and an alternative heuristic solution method reported in the literature.