Research On Some Optimization Problems And Their Application For Metal Structure Production Process

The metal structure are widely used in engineering machinery, mining machinery, port machinery and other industry machinery. This dissertation involves investigations in lot sizing, parts grouping cutting, job scheduling and manufacturing execution system for solving problems such as delivery tardiness, low nesting efficiency, low utilization rate of sheet, large quantities of WIP (Work In Process), as well as the kitting problem and equipment load imbalance, which are very common in the metal structure production process. It is shorten the metal structure production cycle, reduced its production costs and improved its manufacturing management level, laid the foundation for digital manufacturing of metal structures by studying the optimization method for the metal structure production process.Firstly, in study of the method for lot sizing, according to the characteristics of the metal structure manufacturing systems and uncertain disturbance of manufacturing costs and demand, a robust optimization method is proposed to develop the lot size of metal structure. A robust model is first constructed based on the original lot size optimization model.It is then transformed into a linear programming model and solved by the simplex method using LINGO mathematical optimization software. By using this method, the order tardiness and the production costs are reduced. Then the validity of the method is proved by a case analysis.Secondly, in study of algorithms for parts grouping cutting problem, a parts groping optimization method is developed for the large-scale parts integrated cutting stock problem’s characteristics of the metal structure manufacturing. All of cutting parts are first divided into several groups according to the part’s material and processing. Then a feature characteristic matrix for parts’ shap is constructed. A mapping relation between part grouping and the feature characteristic matrix is developed using an artificial neural network algorithm. Finally, the parts are further grouped based on this mapping relationship. By using this method, the number of work-in-process is reduced and structural member with poor complete sets is improved, while the layout efficiency and material utilization contradictory is cushioned.Thirdly, in study of optimization method for job scheduling problem with a variety of cutting patterns, a mathematical model for multi-objective job scheduling problem with process constraints is proposed and an improved hierarchical genetic algorithm (ant colony-hierarchical genetic algorithm) is developed for the better solution. A set of good cutting pattens are chose by ant colony algorithm, and then an optimum solution for job sequencing and machine selection is obtained using the hierarchical genetic algorithm. By using this method, the production cycle is shorten, the number of work-in-process is reduced, and the equipment utilization is improved. The validity of the method is proved by a case analysis.Fourthly, in study of the process management method and application for metal structure, some methods are proposed for data acquisition, production schedule control, WIP management and quality control. As a result, the real-time monitoring of the production process is realized and production plans are executed correctly. In addition, a metal structure manufacturing operation optimization platform is developed base on the above theory and methods. The feasibility and practicability of the methods mentioned in this dissertation are proved through the successful implement in a metal structures enterprise.Lastly, the research work is concluded and the future research efforts are proposed.