In recent years, requirements for artifacts are becoming increasingly complex and stringent. To aid design these more complicated artifacts, co-design is becoming popular; it is a method that uses knowledge gained from involving different stakeholders to tackle these challenges. One such method, the pattern language, is effective in promoting the exchange and sharing of knowledge in co-design, it faces the challenge of extracting and formalizing the knowledge possessed by each stakeholder during its creation. This paper proposes a method to support the construction of pattern languages by applying information processing with generative AI to the available information resources. Furthermore, as the result of verifying the effectiveness of the proposed method using a failure knowledge database and generative AI, it was confirmed that knowledge in the form of “situation, problem, solution” can be output from the available information resources. The result is expected to help the construction of pattern languages and enable the widespread use of this tool.
KAWAGUCHI et al. (Wed,) studied this question.