A Method For Product Personalized Design Based On Prospect Theory And Its Support System

With the continuous improvement of social productive forces and people's material life is extremely rich,customers no longer just satisfied with the function to meet their needs products,but also the appearance of the product shape,style,etc.have their own unique color requirements,in an increasingly business environment,we must recognize the importance of product design.How to rapidly and accurately response to customer demand is the key point in product personalized design.User participation in the product design process will be a very effective solution.Through human-computer interaction,interactive genetic algorithm(IGA)emphasizing on collaborative interaction and user participation can achieve that perfectly.However,IGA widely adopted now neglects the impact of decision maker's psychological changes on decision-making behavior.It's hard to capture users' demands and grasp their preferences,generating slow convergence speed and users' significant fatigue problems.Thus,an interactive genetic algorithm based on prospect theory improved with interval reference(IGA-PTIR)is proposed in the paper.In IGA-PTIR,an evolutionary individual's fitness value is represented by prospect value of prospect theory in order to correct individual fitness deviation caused by ambiguity and uncertainty of user cognition.According to the feature of user's evaluation on individuals with reference to the individuals evaluated,we also introduce interval reference into prospect theory and put forward the concept of interval prospect value,which ensures interval prospect value be closer to real fitness.Then an interval number sorting strategy based on dominance degree is adopted to rank individuals in the same generation and elite individuals are selected,which makes it possible to perform subsequent operations of crossover and mutation.In the paper,IGA-PTIR is applied into automobile wheel hub design system and compared with traditional interactive genetic algorithms(TIGA).Experimental results indicate that the proposed algorithm can improve convergence speed and ease user fatigue to a large extent.