Abstract
Modularity is an approach to manage the design, assembly, maintenance and upgrading of product-systems by partitioning and assigning elements of a product to simpler subsystems according to a planned architecture. Manual guidelines performed with functional-flow diagrams suggest possible modules that have been demonstrated in past products, but using them still leaves it to the designer to choose which guidelines make sense in a certain architecture. This constitutes an opportunity for a designer to take other constraints and objectives into account. With large complex systems, the number of alternative groupings of elements into modular chunks becomes exponentially large and some form of automation would be beneficial to accomplish this task. Clustering algorithms using the design structure matrix (DSM) representation search the space of alternative relative positioning of elements and present one ideal outcome ordering which “optimizes” a particular modularity metric. Such approaches partition the elements in an unconstrained manner and according a single unified modularity metric. Yet, typical complex products are often subjected to constraints which invalidate the unconstrained optimization and objectives that are in trade-off with each other. Here I propose, implement and demonstrate a methodology along with essential computer algorithms for modularization of product systems under constraints and also different objectives. Constraints considered in this thesis are constraints associated with external force fields including electric, magnetic, or pressure fields that constrain some functions to perform or not perform in different regions of the field. In this thesis Objectives discussed in this thesis include connectivity-based and similarity-based objectives. Connectivity based-objectives measure internal integrity of modules and their independence from other modules and similarity-based objectives that are concern about placing components in association with similar product properties in the similar modules. A set of guidelines are developed to incorporate field-related constraints into product modularization and develop an algorithmic approach for enforcing these guidelines into constrained DSM clustering algorithms. I also show trade-off between different connectivity-based and similarity-based objectives and suggest an approach for generating an architecture that is optimal according to multiple objectives of interest. Finally, I demonstrate application of constrained clustering algorithms in case of practical project of SUTD MRI Man-portable Imager and show how computer-generated and human-generated architectures differ in terms of different objectives. Therefore, this PhD research has demonstrated that multiple objectives and constrains can be incorporated to product modularization based on DSM clustering algorithms and including these objectives and constraints makes computer-generated architectures of more practical value for product designers.