|
With the economic development of globalization, the increasingly fierce market competition, China's equipment manufacturing industry is facing enormous challenges. In order to better adapt to the international situation, improve the technological content of China-made machinery and equipment, has won domestic and foreign markets, the paper CXT hoist series as a carrier of its modular design approach, combined with the results of previous studies, presented own set of modular design system. Modular thinking in the equipment manufacturing industry prospects, the future will certainly be vigorously promoted. Modular design method as a new modern design methods and ideas, in the form of things pattern analysis, optimization and system decomposition mechanism, restructuring, coordination, etc., has its own unique performance, so the theoretical system of modular design research is particularly important. Modular design includes module division and rational evaluation, module design, module management, etc., and module division and reasonable evaluation is based on a modular design and implementation of the key. This article focuses on methods and evaluation module division, while the modular design and module management also do some study. This article will divide and reasonableness of the evaluation module is divided into weight indicator to determine the original clustering as well as sub-module program evaluation module division of three parts. For module division, the first is to determine the division of indicators or based on existing literature through research, combined with the characteristics of CXT hoist, this article is meant to give the sub-module functional relevance, geometric correlation between physical relevance such as the module division basis. Therefore, the module division before determining the weights of each correlation is particularly important, this paper is the analytic hierarchy process, when considering the impact on modular division of the various factors to establish AHP model tree, obtained by calculating the correlation weights. Weight determination will be carried out after initial sub-module of the correlation matrix for the establishment and calculation, and then come to a variety of classification scheme. This paper uses a fuzzy cluster analysis. The initial sub-module through the correlation between the functional, geometric correlations, physical correlation obtained correlation matrix, according to different thresholds for clustering, and draw dynamic clustering map, get clustering results. Obtained after clustering module division scheme is not unique, for a variety of module division plan, the paper uses gray relational analysis method is preferred partitioning scheme.
|