A Combinatorial Approach to Tradespace Exploration of Complex Systems: A CubeSat Case Study

Abstract

This paper presents a novel approach to tradespace exploration for large‐scale and multi‐dimensional tradespace, which reduce the cognitive workload in addressing numerous design options to a manageable level. Our approach enables exploration by classifying subsets of design solutions or designs of interest, which will in turn facilitate and accelerate the decision making process. As the design tradespace for products such as CubeSats is naturally large‐scale and multi‐dimensional, the proposed approach integrates computational data clustering algorithms with multi‐dimensional scaling visualization. In the proposed approach, k‐medoids clustering provides a grouping model where similar designs are grouped; uncovering hidden patterns and features within datasets; and the principal component analysis allow users to visualize design samples on two‐dimensional scatter plots. A case study tradespace consisting of 1200+ CubeSat design alternatives is used to demonstrate how this approach supports knowledge discovery of design options through tradespace exploration of multi‐dimensional data.

Decision making for multi‐objective problems: Mean and median metrics

“…The data of this case study and its description have been previously presented in ref. 34. We used the same case study to demonstrate the effectiveness of a completely different method proposed in this investigation.…”

“…In this case study, a CubeSat is modelled as a structure consisting of six COTS components/subsystems. The subsystems are Communication, Power, Solar Panel, ADCS, Command Data Handling, and Antenna, which are actual COTS components from CubeSat Shop 34 . Table 3 shows the specifications for each COTS subsystem's alternative.…”

“…The subsystems are Communica-tion, Power, Solar Panel, ADCS, Command Data Handling, and Antenna, which are actual COTS components from CubeSat Shop. 34 Table 3 shows the specifications for each COTS subsystem's alternative. For example, there are four options for the Communication subsystem, while the remaining five subsystems each have three options.…”

“…Search algorithms for multi‐objective optimization problems (MOP) rely on the incorporation of preferences models where one solution is deemed favorable over another or considered at par during the course of the search 1 . Discrete MOP is widely used in systems engineering applications to perform various trade‐offs for the systems’ functional and non‐functional requirements and for finding the optimal design configuration 2–5 . While in a single objective setting, the search for the preferred optimum solution is simple and more often is a single solution, for a MOP setting without any preference among the objectives, the result is a set of trade‐off solutions 6,7 .…”