Research On Integration Optimization Of Slab Preparation Logistics Scheduling And Hot Rolling Production Scheduling

The hot rolling production scheduling is to select rolling slabs from the candidate slabs in the slab warehouse and determine their rolling sequence,according to the production process requirements.Slab preparation logistics scheduling is to determine the retrieval sequence of rolling slabs and their space allocation in the loading area,according to the rolling sequence priority of hot rolling production.The rolling sequence influences the retrieval sequence of rolling slabs and the space allocation in the loading area.Scientific hot rolling production scheduling and slab preparation logistics scheduling can reduce the residual production capacity of the hot rolling unit,the transition cost of production,the number of shuffles of the slab warehouse,and improve the logistics efficiency.Modeling and solving the problems are challenging because both the production scheduling and the preparation scheduling have the time-space coupling characteristics of sequence and position.Therefore,the modeling and optimization methods of hot rolling production and slab preparation logistics scheduling problems have important scientific research and practical application significance.Taking the hot rolling production stage and the hot rolling slab warehouse logistics as the background,this thesis presents the time-space network modeling method and the improved differential evolution algorithm to solve the problems of slab preparation logistics scheduling based on optimal stacking and the integration optimization of hot rolling slab preparation logistics and production scheduling.The slab preparation logistics scheduling problem is used to determine the minimum number of stacks that are required to be loaded in a stacking sequence with no shuffles.When this minimum number of stacks is satisfied,the integration problem can be transformed into a single stage hot-rolling production planning problem.This thesis focuses on the following aspects:(1)The slab preparation logistics scheduling problem is derived from the slab preparation process of hot rolling warehouse,which is to successively retrieve the slab on the top of the stack and determine the retrieval sequence among different stacks and the stacking positions in the loading area under stacking constraints and retrieval constraints,to minimize the empty stacks in the loading area.A stage based time-space network model is formulated by exploiting the time and space coupling characteristics of slab moving,with the slab moving stage as the time and the storage position as the space,and the nodes in the network represent the storage status in the warehouse by the end of each stage.Due to the difficulty that large-scale problem cannot be solved with the time-space model,a local search algorithm is proposed.Experimental results show that the time-space model can effectively solve small-and medium-scale problems,and the local search algorithm can solve large-scale problems in a short time.(2)The integration of slab preparation logistics scheduling and hot rolling production scheduling is to select rolling slabs from the candidate slabs in the slab warehouse and determine the production and retrieval sequences of rolling slabs and the stacking positions in the loading area,in order to minimize the residual capacity of production,transition costs,the number of shuffles and the number of empty stacks required in the loading area under constraints of hot rolling process,slab stacking,and retrieval.Due to the complex production process requirements,stacking space requirements and the time-space coupling characteristics of the rolling and retrieval sequences of this problem,a multi-dimensional time-space network model is formulated,with the rolling sequence and slab moving stage as the two-dimensional time and the storage position as the space,and the nodes in the network represent the rolling sequence of the slab at the storage position in the warehouse by the end of each stage.Experimental results by the optimization software CPLEX show that the time-space model can effectively solve the small-scale integration problem,and the integrated optimization scheduling mode can effectively reduce the number of shuffles and the total costs compared with the traditional independent optimization mode.(3)Due to the difficulty of the time-space model to solve the large-scale integration problem,an improved differential evolution algorithm is designed to solve the problem.In order to improve the quality of initial population,a two-stage heuristic algorithm is proposed to obtain partial initial solutions.In the first stage,the slabs and rolling sequence are determined.For the second stage,the results obtained in the first stage are taken as input data to determine the retrieval sequence of the slabs and their stacking positions in the loading area.Time-space modeling based solution method is proposed to solve small-scale problems,and a heuristic algorithm is designed to solve large-scale problems.In order to further improve the quality of the solution,According to the characteristics of the problem,adaptive strategy and mutation strategy based on external archives are designed.Numerical experiments show that the improved differential evolution algorithm can effectively solve the integration problem.(4)Based on the proposed time-space model and improved differential evolution algorithm,a decision support system for the integration of slab preparation logistics scheduling and hot rolling production scheduling is designed and developed.The main functions include data information management,candidate slabs pooling,planning,violation checking,and manual modification.The system can realize the hot rolling scheduling,and improve the logistics efficiency of slab warehouse and reduce the logistics cost.