Changeability assessment in systems during early design

Abstract

Increased stakeholder expectations and tougher global competition push the manufacturing industries to develop more complex solutions, introducing numerous uncertainties in design decision-making. Incorporating changeability is one way to systematically deal with uncertainty, which requires dealing with the identification, valuation, and quantification of changeability during the early design stages. Besides, when considering a system-of-systems (SoS) problem, it necessitates optimizing the portfolio of options within the SoS while considering the transition “in” and “of” the system combinatorially. In this light, this paper presents the Changeability Assessment in Systems during Early Design (CASED) method for development teams to create value-robust systems in the face of uncertainty. The CASED method utilizes epochs and eras to represent uncertainty and analyzes the impact of changing system configurations and control policies on overall system value. Considering transitions “in” and “of” the system involves solving a linear assignment problem to minimize the switch costs. A dynamic programming procedure for an era finds the avenues when change shall be exercised. Iterating through several eras, a map of the expected mean value and expected standard deviation is generated for all designs, which serves as a guide for decisions concerning changeability. The applicability of the method is further confirmed through a fleet of haulers operating in a quarry.