A dynamic constraint satisfaction approach for configuring structural products under mass customization

摘要

Configuring structured products poses new challenges to the solving technologies for product configuration. This paper presents a novel and direct approach to encoding configuration models into the Dynamic Constraint Satisfaction Problems (DCSP). In the presented approach, components are encoded as DCSP variables while structural relationships are represented as DCSP activity constraints. Furthermore, the configuration constraints such as the requisition and exclusion constraints are treated as DCSP compatibility constraints, which allow a low-level component to join in the solving process only after its high-level component is selected in the configuration. The presented method allows a more compact encoding representation, compared to CSP and generative CSP. Experimental study shows that the presented DCSP encoding approach makes a significant improvement in the performance of product configuration.