The Multi-objective Optimization Of Hot-rolling Batch Planning And Scheduling For Round Steel Production

As an important part of the iron production managements, the problem of hot-rolling batch planning and scheduling made critical influence on production efficiency, economic benefits and product qualities. Existing researches were mostly focused on products with wide application, such as steel plate, steel strip and steel tube, et al.. However, the corresponding researches on the round steel, which is also an important steel product, were rare. For the special processing technology that different from other steel products, existing theories and methods cannot used for hot-rolling production planning and scheduling in round steel production. Therefore, in this dissertation, the hot-rolling batch planning and scheduling based on the round steel processing technology is studied, and the focus is on question characters and approximation algorithms. The main work and contributions of this dissertation are summarized as follows:(1)For the batch planning problem of round steel based on order grouping, considering the relationship between production and transportation, an order grouping method with transport optimization is proposed. Following the hot-rolling processing technologies, order properties are digitized, and a batch planning model with minimizing the differences between orders is proposed. With analysis for characteristics of the model, an algorithm based on order properties is designed. The solving process is divided into two phases with defining the two batching principles (same property principle and similar property principle):firstly, initial rolling unit is formed with the same property principle, and then, further split on the initial unit is implemented with similar property principle. Considering the transportation optimization, an improved k-means algorithm is proposed to solve the model. simulation experiment shows that the two-stage algorithm can achieve a perfect batch planning results.(2) For general hot-rolling batch scheduling problem of round steel, considering the production process program constraints and production integration objectives, a multi-objective integer programming model is built with the objectives to minimize the makespan, the setup costs and earliness\lateness of orders, and an improved strength Pareto evolutionary algorithm based on population repair(PR-SPEA2) is proposed to solve the model. With the consideration of the discreteness of variables in the model, encoding rule and genetic evolution method are designed to strengthen the population optimizing capacity. For the multi-objective properties of the model, a fitness assignment based on calculating processing time of batches is designed to evaluate the individuals, and the environmental selection is also used for updating the archive. Some rules are proposed to repair unreasonable solutions emerged in the genetic progress of the population. In the end, experiment with actual production data is designed to test the feasible and effective of model and algorithm.(3) For the hot-rolling batch scheduling problem of round steel with machine maintenance, considering the influence on the continuous rolling caused by machine maintenance, a multi-objective integer programming model is built with the objectives to minimize the idle time between batch producing and machine maintenance, the setup times and earliness\lateness of orders. With character of the problem, a modified Non-dominated Sorting Genetic Algorithm Ⅱ (NSGA-Ⅱ) is proposed to solve the model. With the consideration of the constraints in the model, initial solution is generated by a modified NEH algorithm, a fitness function is designed with the thought of penalty function, and some rules are proposed to repair unreasonable solutions. In addition, the genetic methods are designed for the evolution progress, and a local search algorithm based on limited search scope is embedded in the iteration to avoid a local optimum. Numerical results demonstrate that the performance of the algorithm is remarkable in both solution quality and solving efficiency.(4) For hot-rolling batch scheduling problem of round steel with flexible maintenance planning, a multi-objective optimization model, which aims at minimizing the makespan and the earliness\lateness of orders, is built to obtain an effective cooperative scheduling with machine maintenance and batch production. With the consideration on the featuring of the model, an improved multi-objective particle swarm optimization algorithm is proposed. In the proposed algorithm, an insertion algorithm based on fitness assignment with chaos weighting is designed to generate the initial solution. According to the constraints in the model, some rules are proposed to repair unreasonable solutions. With the elitist strategy, advanced individuals are preserved in evolution process, and the extremums for every individual’s updating are also selected from elite set. In addition, with considering the discrete character of variables, genetic operators are introduced to update particles. The feasibility and effectiveness of proposed model and algorithm are verified by computational tests with practical data.