An Interval-based Constraint Satisfaction (IBCS) Method for Decentralized, Collaborative Multifunctional Design

Set-based design has been proposed as a strategy for multifunctional design problems where stakeholders from different disciplines strive to achieve domain-specific objectives while sharing a set of design variables. This strategy involves communicating information about sets of alternatives in contrast to communicating information about a single alternative at a time. The strategy has been developed for collaborative CAD and for selection among design alternatives during conceptual design, it has not been implemented as a computational method for decentralized collaborative multi-objective design problems. In this article, we address this research gap by presenting an IntervalBased Constraint Satisfaction (IBCS) Method for decentralized, collaborative multifunctional design. The method is based on transforming a decentralized multifunctional design problem into a constraint satisfaction problem by using non-cooperative game theoretic protocols. The resulting constraint satisfaction problem is then solved using interval-based consistency techniques. A non-cooperative game theory protocol is utilized in this method because it reflects the level of information exchange possible in a distributed environment. Central to this protocol is the representation of a Rational Reaction Set (RRS) that encapsulates a designer’s decision-making strategy as a constraint in the design space. An intersection of all designers’ RRSs represents a solution to the overall multifunctional design problem. We use interval-based consistency techniques, specifically box consistency, to sequentially eliminate regions of design space that do not satisfy the individual RRSs, thereby progressively narrowing the design space in order to reduce computational complexity in arriving at a solution. This method stands in marked contrast to the successive consideration of single solution points, as emphasized in existing multifunctional design methods. The key advantages of the proposed method are: (a) gradual reduction of design freedom and (b) non-divergence of solutions. The method is illustrated using two sample scenarios – the solution of a decision problem with quadratic objectives and the design of multifunctional Linear Cellular Alloys (LCAs).