DCR-based causal design knowledge evaluation method and system for future CAD applications

This paper presents a new causal design knowledge evaluation method and system for future CAD applications. Current product development processes still include unintended feedback due to insufficient product design knowledge. Previous research on design knowledge support system focuses on search by matching keywords and file names, or search by specific indices, which has various drawbacks. Furthermore, current CAD systems need manual input to incorporate the designer’s knowledge. To systematize the knowledge management process for the next-generation CAD systems, a prerequisite is to capture ever-evolving causal design knowledge. In this paper, we present a new causal knowledge network evaluation method, which has not been well addressed in design knowledge support system research. For the network evaluation, we present a degree of causal representation (DCR)-based knowledge network evaluation method. In this method, causality and network connectivity are used for the causal knowledge network with weighted vertices and weighted network connectivity for a network with weighted edges. To validate the proposed method, this evaluation method has been compared with structural measures. Finally, the causal design knowledge evaluation system, KNOES, is implemented and tested with a new valve design scenario.

Research highlights
► A new causal design knowledge evaluation method and system for future CAD applications is developed.
► This new CKN evaluation method compares multiple design knowledge networks using DCR and supports design decision.
► Causal design knowledge networks and their evaluation results can be used to predict the effects of design change and sensitivity analysis for future potential failures.