Name
Abstract | ||
|---|---|---|
Product design is an important part of the manufacturing system from a macro point of view. In the design process, creative design is one of the important factors to survive the fierce competition. During creative design process, designers are the most critical component. However, design fixation often negatively influences the design outcomes. Many researches from cognitive science and design science reveal that the presentation of outer information can alleviate design fixation effectively. Among different kinds of outer information, language terms are proved to be effective. This work presents a method of automatically generating language terms as design inspirations based on 500,000 granted patents. This method adopts evolutionary computation as the fundamental algorithm to retrieve language terms from a vocabulary base. To implement the algorithm, the vocabulary is first encoded by high dimensional vectors through word embedding, which is a three layers neural network. Further, two metrics for measuring the inspiration potential of language terms are defined in a computable manner. This work also conducts experiments to validate the method, and the experimental results show that (1) the algorithm is efficient and has the potential to be extended to larger vocabulary; (2) the generated design inspirations have a positive influence on the design outcomes. |
Year | DOI | Venue |
|---|---|---|
2019 | 10.1007/s10845-017-1347-x | Journal of Intelligent Manufacturing |
Keywords | Field | DocType |
Design fixation, Design inspiration, Evolutionary computation, Word embedding | Computer science,Theoretical computer science,Engineering design process,Artificial intelligence,Word embedding,Product design,Generative Design,Design language,Mathematical optimization,Design science,Computer-automated design,Vocabulary,Machine learning | Journal |
Volume | Issue | ISSN |
30.0 | 4.0 | 1572-8145 |
Citations | PageRank | References |
1 | 0.35 | 13 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Jia Hao | 1 | 9 | 3.67 |
| Yongjia Zhou | 2 | 1 | 0.35 |
| Qiangfu Zhao | 3 | 214 | 62.36 |
| Qing Xue | 4 | 38 | 5.71 |