Study On The Group Technology Of Revolving Body Components Based On Genetic Algorithm

Group technology has been used extensively in the manufacturing field for decades after it generated. But in its developing process, we can make out that if only consider group technology as a singleness method, and ignore the understand of its aboard philosophic theory, group technology maybe loss of self-dominance in the drastic market competition. So, we can combine group technology classical pattern to the current multi-disciplinary advanced technology, and form a new research direction. Genetic algorithm is an fresh subject in recent years, It is a search algorithm based on the mechanics of natural selection and natural genetics. It is widely used in many kinds of fields because of its less-dependency of optimization problem, simplicity robustness and implicit parallelism. Grouping problem is combinatorial optimization problem which is effective to solve by genetic algorithm. So this thesis connects group technology and genetic algorithm, presents a components grouping method based on genetic algorithm.This thesis sets up a revolving body components grouping system based on genetic algorithm, carry out grouping of the revolving body components. First, by analyzing the representative revolving body components, this thesis frames a characteristic group form for revolving body components, and picks-up the characteristic value of the components.Second, for different aim of group, we should think synthetically about the different of the effect of each component characteristic on group result and separate them into three, four, five, six groups, respectively. Simultaneity, we can count similarity coefficient of the components.Finally, this thesis provides its mathematics model and proposes a solution method based on genetic algorithm. In order to satisfy complex constraints of mathematics model and gives matrix code of chromosome and the corresponding operators are given, and also achieves the algorithm simulation by using the MATLAB programming. Results indicate that: when maximal sum of within cluster similarity coefficient is taken to be performance criterion, the above method is accurate. Simultaneously, using the method of both traditional coding...