An interactive genetic algorithm with the interval arithmetic based on hesitation and its application to achieve customer collaborative product configuration design

• An interactive genetic algorithm with the interval arithmetic based on hesitation is proposed.
• An interval number derived from user's evaluation time is adopt to express an individual's fitness.
• The interval probability dominant strategy is proposed to compare advantages and disadvantages of evolutionary individuals.
• The proposed algorithm is applied to the design system of the car console configuration.

Complex product configuration design requires rapid and accurate response to customers’ demand. The participation of customers in product design will be a very effective solution to achieve this. The traditional interactive genetic algorithm (IGA) can solve the above problem to some extent by a computer-aided user interface. However, it is difficult to adopt an accurate number to express an individual's fitness because the customers’ cognition of evolutionary population is uncertain, and to solve the users’ fatigue problem in IGA. Thus, an interactive genetic algorithm with interval individual fitness based on hesitancy (IGA-HIIF) is proposed in this paper. In IGA-HIIF, the interval number derived from users’ evaluation time is adopted to express an individual's fitness, and the evolutionary individuals are compared according to the interval probability dominant strategy proposed in this paper. Then, the genetic operations are applied to generate offspring population and the evolutionary process doesn’t stop until it meets the termination conditions of the evolution or user manually terminates the evolution process. The IGA-HIIF is applied into the design system of the car console configuration, and compared to the other two kinds of IGA. The extensive experiment results are provided to demonstrate that our proposed algorithm is correct and efficient.

As users have limited knowledge about the population when they are evaluating evolutionary individual, and it is hard for them to give the precise fitness for the evolutionary individual, interactive genetic algorithm with interval individual fitness based on hesitancy time(IGA-HIIF) is proposed in the paper, in which a fuzzy number of interval is used to show users’ evaluation for the evolutionary individual, and the length of hesitancy time when they evaluate to indicate the width of the interval. In this part, the key theory about how to use interval based on hesitancy time to indicate evolutionary individual, use interval fitness to compare and judge them is discussed in details because it directly influences on the select operation of the proposed method.Figure optionsDownload as PowerPoint slide