When the amount of data to be reviewed is large and the properties of the material are complex, it is difficult to make a rational decision in selecting the optimal material. Therefore, in this study, we tried to develop an optimization model that comprehensively considers user requirements, performance and economic feasibility of materials for selecting materials with low emission of indoor air pollutants. To this end, a database was constructed considering the economic feasibility by applying the concept of LCC (Life Cycle Cost) and presenting price range options that can be selected by the user. A genetic algorithm was used to construct a model to derive a material plan that could achieve the target score while satisfying economic feasibility and user requirements. As a result of model verification and verification cases, materials were selected only within the range according to the price range option and user selection criteria for each space and part. The efficiency and effectiveness of this model were confirmed. In this study, reliable results can be presented by presenting a model that can automatically select an algorithm for the optimal preferred material selection problem that is difficult for humans to solve cognitively with database construction and user selection information. Since it can be used in other fields, scalability and usability of this model are expected. In addition, it helps user to reduce the time of the material selection process and the price of materials is also considered, so that it is expected to help improve the economic feasibility of overall construction.

Recent advancements in data mining techniques showcase expansive and effective insights. Yet, these insights often remain incomplete. In situations demanding optimal results, such incomplete knowledge from various agents falls short. This paper delves into the role of mobile agents engaging in data mining within a multi-agent environment. Each agent is tailored with distinct goals, mining data in line with its identified problem set. Mobile agents understand insights from relevant agents equipped with specialized data knowledge. Outcomes from diverse agents converge, serving as foundational data for mobile agents. This integration is facilitated by integrating adaptive automata and genetic algorithms, enhancing the mobile agent’s expertise on a particular task.