• CN: 11-2187/TH
  • ISSN: 0577-6686

Journal of Mechanical Engineering ›› 2017, Vol. 53 ›› Issue (15): 73-80.doi: 10.3901/JME.2017.15.073

Previous Articles     Next Articles

An Evolutionary Computation Method of Crowdsourcing Ideation that Integrates the Balanced-exploration Pattern

XIANG Wei1,2, SUN Lingyun1,2, XIA Suochen2, XU Jiang3   

  1. 1. State Key Lab of CAD & CG, Zhejiang University, Hangzhou 310058;
    2. Modern Industrial Design Institute, Zhejiang University, Hangzhou 310017;
    3. College of Design and Innovation, Tongji University, Shanghai 200092
  • Received:2016-08-29 Revised:2016-11-03 Online:2017-08-05 Published:2017-08-05

Abstract: Crowdsourcing ideation recruits crowds to complete design tasks such as product design and graphic design, acting as an important method of generating design solutions. However, crowdsourcing participants are amateurs and hard to organize. An evolutionary computation method of crowdsourcing ideation that integrates the balanced-exploration pattern is proposed in order to combine experts' competence with computers' computing capacity. This method holds the process of crowdsourcing ideation based on the study of design experts' behavior so that the crowds act as design experts, thus enhancing the quality of design solutions. Specifically, the balanced-exploration pattern summarized from the experts' behavior, including "constant information seeking", "structural exploration", "mix of breadth-first and depth-first modes of solution development", and "mix of solution-exploration and solution-review" is incorporated, the evolutionary fitness of design solutions is calculated, the coding, crossover, and mutation operations are refined, thus orienting the evolution of design solutions. To verify this method, an experiment of crowdsourcing ideation has been conducted during which 252 participants produces three generation of design solutions. This case validates that the crowdsourcing ideation method increases the diversity of design solutions and improves the quality of design ideas.

Key words: behavior pattern, crowdsourcing ideation, design cognition, evolutionary computation

CLC Number: