Scheme Evaluation Method Research Based On Tacit Knowledge

A scientific scheme evaluation method is a theoretical basis for the scheme evaluation module of the product intelligent design system. Because the evaluation of the design scheme needs to consider multiple evaluation indexes, if not considering the relationship among the evaluation indexes, and simply assuming that the index is independent of each other, it is bound to affect the objectivity and accuracy of the evaluation results. But there is no consideration of the problem of the relationship among the evaluation indexes. The reason is that there is no suitable modeling method to express the comprehensive influence relationship among the evaluation indexes, on the other hand, there is no fusion mechanism which can combine the relationship with existing scheme evaluation method, so that it can not construct a complete scheme evaluation method.Therefore, this thesis introduces the modeling method of tacit knowledge into the field of scheme evaluation, and studies the scheme evaluation method research based on tacit knowledge. First, through the comparative analysis from definition and characteristics, it is demonstrated that the influence relationship among the evaluation indexes is a kind of tacit knowledge, which can be obtained and expressed by the relevant model of tacit knowledge. Then, according to the specific areas of scheme evaluation, the fuzzy cognitive map is used as the main model to obtain and express the expert’s tacit knowledge on the influence relationship among the evaluation indexes. On the basis of this, the fuzzy decision mechanism is constructed, which aims to transform the fuzzy cognitive map into the index weight, so that it can be combined with existing scheme evaluation method. And then, the grey relational analysis is chosen to combine with the fuzzy decision mechanism, so that it will be a complete scheme evaluation method which can deal with the problem of the influence relationship among the evaluation indexe, namely the FDM-GRA scheme evaluation method. The method was verified by the example of the design scheme evaluation of the small arms, and the application results proved the feasibility and effectiveness of the proposed method.At the same time, taking into account about the real needs about fusion of a number of expert knowledgethe experience, and exploiting historical data, this thesis finally introduces the fuzzy cognitive map learning algorithm RCGA to improve the FDM-GRA scheme evaluation method, namely a scheme evaluation method based on historical data learning. This method can optimize the structure of the fuzzy cognitive map by the study of historical data, which makes the evaluation method more objective and reliable.