Facilitating Engineers Abilities to Solve Inventive Problems Using CBR and Semantic Similarity - Automated Invention for Smart Industries18th International TRIZ Future Conference, TFC 2018 Access content directly
Conference Papers Year : 2018

Facilitating Engineers Abilities to Solve Inventive Problems Using CBR and Semantic Similarity

Abstract

Our industry currently undergoes a period of important changes. The era of computerization implies to companies to change not only through their organization, but also in automating as much as possible their internal processes. Our research focuses on the computerization of the problem-solution couple when facing inventive situations in R&D. The method used is based on Case-Based Reasoning (CBR) that has already been proven to be useful in routine design. On the other hand, CBR is hardly used in inventive situations because the latter require reasoning outside the circle of knowledge recorded in a database. Our proposal consists in coupling CBR with semantic similarity algorithms. The aim is to resolve a new problem based on its semantic similarity with the old problems. Then the old solution can be adapted to solve the new problem. We postulate that a multidisciplinary case base sufficiently populated of multidisciplinary problem-solution couples is likely to considerably improve the performance of R&D engineers to solve inventive problems. This being possible by bringing them alternative solutions based on the semantically similar problems, which are distant from their field of origin. In this way, we provide the possibility to enhance the inventiveness of solution. This type of reasoning, largely inspired by the TRIZ theory, is the subject of this paper. The methodology, the experiments and the conclusions that we develop here validate that this type of approach produces the claimed effects on designers although limited to the context where it has been conducted.
Fichier principal
Vignette du fichier
474537_1_En_17_Chapter.pdf (602.59 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-02279761 , version 1 (05-09-2019)

Licence

Attribution

Identifiers

Cite

Pei Zhang, Denis Cavallucci, Zhonghang Bai, Cecilia Zanni-Merk. Facilitating Engineers Abilities to Solve Inventive Problems Using CBR and Semantic Similarity. 18th TRIZ Future Conference (TFC), Oct 2018, Strasbourg, France. pp.204-212, ⟨10.1007/978-3-030-02456-7_17⟩. ⟨hal-02279761⟩
46 View
93 Download

Altmetric

Share

Gmail Facebook X LinkedIn More