Improved variable neighbourhood search for integrated tundish planning in primary steelmaking processes

Abstract

Production planning (or product design) in the steel industry needs specific, sophisticated procedures in order to guarantee competitive plant performance. This paper describes an integrated tundish planning problem, considering the steelmaking-continuous casting-hot rolling and other downstream integrated technical constraints, and a multi-objective optimisation model is proposed with the objective to optimise the number of tundish, the additional cost of technical operations and the throughput balance to each flow. Also, instead of using traditional metaheuristic algorithm or artificial intelligence (AI)-based heuristic approaches, this paper develops two new approaches, the improved variable neighbourhood descent (IVND) search method and improved reduced variable neighbourhood search (IRVNS) method, by introducing the iterated local search into local search to the problem described above. The performance of IVND and IRVNS are analysed based on changing the number of local iteration and weights of objective function, these two algorithms are also compared with tabu search(TS) and heuristic method based on numeral analysis of the actual data, and the results show that the model and algorithm are feasible and efficient.

Keywords:

• integrated tundish planning
• variable neighbourhood descend search
• reduced variable neighbourhood search
• multi-objective optimisation

Acknowledgements

This work is supported by the National Natural Science Foundation of China under Grant No. 71071028, No. 71021061, No. 70931001 and No. 61070162, Specialised Research Fund for the Doctoral Program of Higher Education under Grant No. 20070145017 and No. 20100042110025, the Fundamental Research Funds for the Central Universities under Grant No. N090504006, No. N100604021 and No. N090504003, the Hong Kong Polytechnic University funding (Project No. G-YG81)