Research On Multi-objective Reasoning Technique Of Customer Requirements Based On Game Theory

For dynamic, incomplete requirement information in the personalized product customization under the background of big data, the enterprise pays more and more attention to maximize meet customer requirement, and achieve the optimal configuration of product. It has become an urgent problem to be solved that how to realize real-time guidance of dynamic customer requirement, while balanced customer satisfaction and enterprise product cost in product configuration. Processing customer requirement and using reasonable requirement guide mechanism, then using reasonable product configuration reasoning method that combined with computer technology, is an important way to solve this problem, and also the key to achieve mass customization. The issues related to customer requirement were studied, the ant colony algorithm mechanism and game theory were introduced, the guidance mechanism of dynamic customer requirement based on the ant colony algorithm and the multi-objective reasoning model of product configuration based on Bayes-Nash equilibrium were established in this paper. Meanwhile, the solution to Nash equilibrium of product configuration and decision-making optimization of configuration scheme were researched deeply in this paper. Details are as follows:1) The way of gaining customer requirement information was studied by its characteristic. Customer requirement was normalized, customer requirement was divided into static and dynamic requirements, requirement nodes interaction relation structure which has clear multi-domain function level was established and weighted processing. Therefore, the fuzzy, incomplete customer requirement information was turned into clear, available customer requirement vector.2) Aim at incomplete customer dynamic requirement, in order to ensure the correct expression of customer requirement and feasibility of the product configuration, the ant colony algorithm mechanism was introduced, and the guiding mechanism of customer dynamic requirement based on ant colony algorithm was established. Product configurable node and functional requirement node were proposed, the mapping relationship between these nodes was studied, and the model base on product configurable nodes was established. The behavior mechanism of ant colony algorithm was studied, the real-time guidance for customer dynamic requirement information was achieved.3) For balancing customer satisfaction and enterprise product cost, the equilibrium model among product multi-domain nodes was established based on Bayes-Nash equilibrium. Enterprise and customer were used as decision makers, enterprise product cost and customer satisfaction were used as game payoff function, and strategy set of the game was determined by seeking the maximum similarity of structure, performance and cost domain. Finally, Nash equilibrium test standard in product configuration was given, and making a decision analysis for product configuration schemes.4) Nash equilibrium solution in product configuration was proposed, and simulated annealing algorithm (SA) was improved, real-time solution to the Nash equilibrium point of multi-objective reasoning in product configuration was achieved.5) In order to solve the problem of no Nash equilibrium in product configuration multi-objective reasoning so as to realize the optimization of product configuration decisions, the decision optimization method based on variant design was proposed. In order to solve the problem of empty path during the guiding process so as to realize the optimization of requirement guide decision-making, the decision optimization method of customer requirement guiding based on Pareto was proposed.