Interactive analysis of product development experiments using On-line Analytical Mining

• We propose an On-line Analytical Mining based on product data analysis technique.
• We propose a comprehensive OLAM process based on the information system architecture.
• An experiment result shows the feasibility of the proposed architecture and process.
• The architecture and operations are important in ambiguous product development processes.

This paper proposes an On-line Analytical Mining (OLAM) based product data analysis technique that analyzes product data in Product Data Management (PDM) databases. It aims to understand and evaluate product development processes indirectly through product data analysis. In this approach, OLAM integrates On-line Analytical Processing (OLAP) with data mining, allowing it to support flexible and interactive OLAP operations to find product data patterns in intangible and uncertain product development processes. This integration also allows efficient preparation of product data for various data analysis and data mining processes. In this paper, a product data analysis process and its information system architecture are proposed, and the prototype system is implemented to show the efficiency and feasibility of the approach. In addition, the proposed approach is applied to clarify critical failure factors in product development experiments to show that the proposed approach can help engineers proactively evaluate and manage in-progress product development processes.