Multiple population search strategy for routing selection and sequence optimization of a supply chain

Abstract

The current paper outlines a framework of a distributed hierarchical model for supply chain planning and scheduling optimization. The framework comprises three main modules: routing and sequence optimization, supply chain virtual clustering and supply chain order scheduling. It is envisaged that the hierarchical model can be used to realize management level strategies, facilitate planning and optimize the detailed operation schedules of various supply chain units in a supply chain. The detailed design of a multiple population search strategy (MPSS) based on genetic algorithm (GA) and tabu search (TS) for routing selection and operation sequence optimization is presented. Using the tabu search, the crossover and mutation rates of GA can be made adaptive to suit different stages of search. The results show that the MPSS is not only able to reach a better solution, but also able to reduce the computational time. The work has also demonstrated the possibility of adopting a hybrid approach that combines the strengths of the tabu search and genetic algorithms for the optimization of routing and sequence in a supply chain in order to achieve management level objectives such as minimizing cost and increasing the level of on-time delivery.