Solving hot rolling scheduling problem by a new population-based extremal optimization algorithm
Kai Sun, Genke Yang, Changchun Pan · 2010
Hot rolling scheduling problem (HRSP) is an important research and development area for both academic and steel industries. In the paper, the problem is formulated as a prize-collecting vehicle routing problem (PCVRP), which considers two major requirements: (a) selecting a subset of slabs from manufacturing slabs to be processed; (b) determining the optimal production sequence under multiple constraints, such as sequence-dependant transition costs, non-execution penalties etc. And then, a new algorithm which combines the extremal optimization (EO) with population evolutionary technique, called PEOA, is proposed to solve the problem. The proposed algorithm is applied to a set of real production data to test its performance and efficiency. The experimental results show that the new PEOA is very effective to solve the problem.