Analogical stimuli retrieval approach based on R-SBF ontology model

ABSTRACT

Analogy-based design is an effective approach for innovative design involving knowledge representation, analogical sources retrieval, mapping between two systems, and adaptation of candidate solutions. The knowledge representation is the first task to determine whether analogy can be implemented successfully. It is, therefore, necessary to have an efficient description method for knowledge representation and information retrieval. Based on the Structure-Behaviour-Function (SBF) model, this paper proposes an R-SBF model to integrate the flow, structure, behaviour and function of design knowledge. Relational types of states are used to describe the information behaviour with relations of ‘and’, ‘or’, ‘not’, ‘sequence’, ‘parallel’ and ‘feedback’. The R-SBF ontology model is constructed using the ontology editing tool Protégé 5.2.0 to link the knowledge representation and analogues retrieval. In the proposed algorithm of computing function similarity, both the semantic similarity and conceptual correlation are considered for the query scalability. The evaluation criteria for feasibility include Recall, Precision, and F-measure. An analogy-aided design innovation software prototype is developed to support the design activity. The improvement of a robot vacuum cleaner is discussed as an example of applications of the proposed method. The solution illustrates the effectiveness of the analogical stimuli to facilitate the design performance.

KEYWORDS:

• Analogy-based design
• innovative design
• knowledge representation
• analogues retrieval
• R-SBF ontology model

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Lizhen Jia http://orcid.org/0000-0002-7184-859X

Qingjin Peng http://orcid.org/0000-0002-9664-5326

Runhua Tan http://orcid.org/0000-0002-6797-8199

Xuehong Zhu http://orcid.org/0000-0003-4824-1869

Additional information

Funding

This paper is supported by the National Natural Science Foundation of China: [grant number 51605135]; National Natural Science Foundation of China: [grant number 51675159].