SUMMARY

Researchers have paid little attention to the fact that most design activities are truly more like redesign. These activities are characterized by an attempt to leverage experience, knowledge, and the capital that has been sunk into the engineering systems that a company already produces. In this dissertation, it is proposed that an approach be developed to aid designers in making decisions in redesign problems when there exist systems to be leveraged and multiple new systems to be created. In addition, strategy is introduced to the problem through the consideration that new systems may not be offered all at once as is often assumed in product family design research. The aim of the designer is then assumed to be the creation through redesign of a series of new systems with desirable and distinct performance levels along with a plan to realize those systems that involves as little redesign effort throughout the life of the family of systems as possible. The proposed approach is based upon the concepts of Constructal Theory and previous work to create methods for the design of mass customized families of products. The existing methods are abstracted and heavily modified through the infusion of compromise Decision Support Problems at all stages of the decision-making process. In addition, two indices are developed to represent considerations unique to redesign as opposed to original design. These indices for redesign effort and commonality value are utilized in the overall objective formulation for the approach. Through a thorough validation process and a large number of redesign scenarios, it is shown that the overall approach proposed can lead the designer towards promising redesign plans involving leveraging of existing systems, but that the constructal-inspired approach in and of itself has certain limitations when applied to redesign.