Bayesian Networks for Set-Based Collaborative Design

Abstract

A set-based approach to collaborative design is presented, in which Bayesian networks are used to represent promising regions of the design space.In collaborative design exploration, complex multilevel design problems are often decomposed into distributed subproblems that are linked by shared or coupled parameters. Collaborating designers often prefer conflicting values for these coupled parameters, resulting in incompatibilities that require substantial iteration to resolve, extending the design process lead time without guarantee of achieving a good design. In the proposed approach to collaborative design, each designer builds a locally developed Bayesian network that represents regions of interest in his design space. Then, these local networks are shared and combined with those of collaborating designers to promote more efficient local design space search that takes into account the interests of one's collaborators. The proposed method has the potential to capture a designer's preferences for arbitrarily shaped and potentially disconnected regions of the design space in order to identify compatible or conflicting preferences between collaborators and to facilitate a compromise if necessary. It also sets the stage for a flexible and concurrent design process with varying degrees of designer involvement that can support different designer strategies such as hill-climbing or region identification. The potential benefits are the capture of expert knowledge for future use as well as conflict identification and resolution. This paper presents an overview of the proposed method as well as an example implementation for the design of an unmanned aerial vehicle.

Set‐based design: The state‐of‐practice and research opportunities

“…From this set, we identified and collected only refereed journal articles and conference papers, reducing the set to the 64 SBD specific documents considered in the final review. We refer to this set of documents as Group A throughout the remainder of this paper 2–65 . We provide the complete listing for all Group A documents in Table 1.…”

“…Decision and tradespace analysis, along with modeling uncertainty applications were the most commonly used analytical methods. On the other end of the spectrum, the categories of network modeling, 10 game theory, 34 and Bayesian methods 41 are each observed only once within the Group A documents.…”

Making design decisions under uncertainties: probabilistic reasoning and robust product design

“…In engineering design, decisions in the early phases of product development often have to be made under high uncertainty regarding the impact on the entire product life cycle. In the related literature, Bayesian networks are often used to model the uncertainty in order to represent the possible solution space (Moullec et al, 2013;Shahan & Seepersad, 2009), to reduce large data sets by integrating them into a model (Hanafy & Elmaraghy, 2011) or generally to represent relationships between sets of variables in the form of probabilities (Ren et al, 2009;Zhu & Deshmukh, 2003).…”

Set-based design: a review and new directions

“…(i) Multi-objective problem/multidisciplinary optimization (Moshaiov & Avigad 2007;Avigad & Moshaiov 2009 (Shahan & Seepersad 2009, 2012Matthews et al 2014) Figure 5 displays the distribution of the 154 relevant publications by categories. Category 1 is very similar to category 2 with the only difference that category 1 focuses on the method itself (e.g., fuzzy set-based approach, interval set propagation, method of imprecision (MoI), etc.)…”