Research On Nesting Optimization Principle And Key Techniques For Generating Equipment Manufacturing

After China' s accession to the World Trade Organization(WTO), generating equipment manufacturing enterprises face more intensive global market competition. In order to win the market, one urgent task is to increase economic income, which can be realized by decreasing production cost. How to find the method to increase the material utility ratio as much as possible is one of key technical methods to meet the requirement of decreasing product cost. In fact, the cutting stock problems with pattern constraints occur across the full spectrum of engineering and are of great significance and urgency for many fields. These kinds of problems belong to Nondeterministic Polynomial Complete (NPC) problem with the highest complexity in math and the research on them are far from enough. So research on the problem is of importance in theory and practice.Based on the analysis of the features, key procedure and research tendency of various algorithms for cutting stock problem, this paper proposes that the increasing of material utility ratio can not be reached by solving cutting stock problem only. A integrated solution should be resolved from both management and application optimization technology in view of system engineering. Management ensures the execution of centralized batch cutting and optimal algorithms are responsible for the supply of the best cutting pattern and decision support information for planning, scheduling and materials departments. According to the idea mentioned above, a optimal nesting system is developed which is integrated in the workshop production information system for the generating equipment manufacturing and some research work is done on optimal nesting algorithms emphatically. The mainThe research work is supported by the project of National 863 Hi-Tech Plan/CIMS Application Engineering under Grants No 863-511-9702-6123.achievements and creative ideas of the paper are as follows:(1) According to the contributions of researchers at home and abroad and requirements of manufacturing enterprises in China, this paper deals with a general model of agile workshop production scheduling system. The model is used in the research and development of information management system and the function model is established for Cutting Workshop, Dong Fang Electrical Machinery Co. Ltd (DFEM).(2) Genetic Algorithm is an important intelligent algorithm. The first step is to code the solution into permutation when GA is used to solve problems. This paper points out that it is efficient to code the string of index of the parts in adopting GA to solve cutting stock problem after analyzing the features and research tendency of GA. This paper presents suitable operators of selection, crossover and mutation for the coding method. These operators are used in the calculation of one-dimensional and rectangular cutting stock problems. The fitness definition functions under different processing condition are discussed. Results of numerical examples show the efficiency and validity of the algorithm.(3) Research on the rectangular packing problem not only deals with the layout problem of a set of 2D rectangular items onto a rectangular object but also plays a important role in the solving of irregular cutting stock problem. This paper establishes the mathematical model, analyzes the procedure and features of several heuristic packing algorithms and presents a " Lowest Horizontal Line" method which meets the BL condition and overcome the shortcomings of other algorithm for some patterns. Numerical examples shows the efficiency of the LHL method. Two meta-heuristic hybrid algorithms are given that integrated LHL method with genetic algorithm and simulating annealing algorithm separately. The coding and decoding methods, key parameters and the definition of fitness function are discussed. The determination of crossover possibility and mutation possibility is shown by experiments. The parameters of cooling schedule for simulating algorithm is a...