Integer programming model and dynamic programming based heuristic algorithm for the heavy plate shuffling problem in the Iron and steel industry

This paper presents the heavy plate shuffling problem (HPSP), which is motivated by the logistics operations of the heavy plate outgoing in the finished product warehouse. The required plates are taken out according to the given retrieval sequence. If a required plate is located under other plates in a stack, shuffling occurs. HPSP determines the restorage spaces for plates to be shuffled while minimizing the total shuffling number and the total crane traveling distance. The problem is formulated as an integer-programming model by considering the practical stacking and shuffling constraints on the heave plate. Furthermore, a two-phase heuristic algorithm based on dynamic programming is proposed to solve it. In the first phase, an initial solution is generated by approximate dynamic programming. And then it will be improved through an iterated local search in the second phase. Compared with the current used approach, computational results show that the proposed new method reduces the total increased shuffling number by 10.18% and the total traveling distance by 3.27% on average, respectively.