A Decision-Theoretic Graphical Model for Collaborative Design on Supply Chains

We propose a decision-theoretic graphical model for collaborative design in a supply chain. The graphical model encodes the uncertain performance of a product resultant from an integrated design distributively. It represents preference of multiple manufacturers and end-users such that a decision-theoretic design is well-defined. We show that these distributed design information can be represented in a multiply sectioned Bayesian network. This result places collaborative design in a formal framework so that it can be subject to rigorous algorithmic study.