Research On The Design Of Product Configuration Based On Interactive Genetic Algorithm With Co-evolution Of Small Populations

In the increasingly fierce market competition environment,customers have become an important resource for product manufacturers to compete for.In order to attract customers and retain them,manufacturers must mine customer preferences accurately and respond to their needs quickly.Interactive genetic algorithm can effectively solve this problem by involving customers directly in product design process.However,since people need to interact frequently with the computer to assign the individual's fitness value,user fatigue problem becomes a bottleneck for the further development and application of the method in practice.Thus,on the basis of interactive genetic algorithm,an interactive genetic algorithm model with fuzzy relative fitness based on co-evolution of small populations(IGA-FRFC)is proposed in the thesis.In IGA-FRFC,the number of evolutionary individuals evaluated by users is decreased by reducing the size of the population,thus alleviating the problem of user fatigue.Meanwhile,the method of population co-evolution is adopted by the algorithm to ensure the diversity of evolutionary individuals,avoiding the local convergence problem of small population evolution.The way of evaluating evolutionary individuals by users is improved and the method of choosing the better with comparison is adopted to determine the relative fitness values of individuals in the algorithm,which can improve the speed of user evaluation of evolutionary individuals effectively and release the burden of user evaluation greatly.In the thesis,the proposed algorithm is applied in the configuration design of the smart phone's appearance and the its performance is compared with that of the traditional interactive genetic algorithm.Experimental results illustrate the feasibility,validity and correctness of the proposed algorithm strongly.The algorithm model proposed in the thesis completes the theoretical framework of complex product configuration based on the existing methods.The achievement of the thesis based on the proposed algorithm can not only be extended to actual production by means of computers but also by small-screen smart mobile terminal devices such as mobile phones and tablets,helping product manufacturers to formulate new product development and design strategies,and the application prospects of the thesis' s achievement of are further expanded.