This dissertation, "Evolutionary Optimization Methods for Mass Customizing Platform Products" by Li, Li, 李麗, was obtained from The University of Hong Kong (Pokfulam, Hong Kong) and is being sold pursuant to Creative Commons: Attribution 3.0 Hong Kong License. The content of this dissertation has not been altered in any way. We have altered the formatting in order to facilitate the ease of printing and reading of the dissertation. All rights not granted by the above license are retained by the author. Abstract: Abstract of thesis entitled Evolutionary Optimization Methods for Mass Customizing Platform Products Submitted by Li Li for the degree of Doctor of Philosophy at The University of Hong Kong in August 2007 Platform product development is an effective approach used by manufacturing companies to realize mass customization. It involves two essential stages: (i) establishing a product platform; and (ii) customizing the platform into product variants. This thesis addresses the second stage in two forms: (a) platform product customization (when the platform is customized into product variants one at a time); and (b) product family design (when the platform is customized into products variants simultaneously as a family). This study uses the concept of generic bill-of-material to represent product platform and instance bill-of-material to represent product variant. Thus, customizing the platform into product variants becomes naturally a process of establishing instance bill-of-materials from a given generic bill-of-material. Platform product customization takes place at the structural and parametric levels in the process. The study proposes to conduct structural optimization and parametric optimization simultaneously in integrated computation models. Three algorithms have accordingly been developed. The first algorithm is a tandem evolutionary algorithm where a genetic algorithm for parametric optimization is nested within a genetic programming for structural optimization. The second algorithm is an interwoven evolutionary algorithm where a genetic programming for structural optimization and a genetic algorithm for parametric optimization interweave with each other to evolve. The third algorithm is a cooperative coevolutionary algorithm where the two evolutionary processes for structural optimization and parametric optimization are evolved cooperatively. Comprehensive experiments are conducted to test the validity and evaluate the effectiveness and efficiency of these three computational methods. Results have shown that they are able to converge with solutions of good quality. The tandem evolutionary algorithm reflects the iterative optimization at the structural and parametric levels intuitively. But its computational efficiency is found to be the lowest of the three methods. The interwoven evolutionary algorithm demonstrates a competitive efficiency, because the interwoven mechanism enables the algorithm to simultaneously explore the solution spaces of the two sub-problems. The cooperative coevolutionary algorithm exhibits the best performance in solving complex problems, while the interwoven evolutionary algorithm is found to be more efficient in solving less complex problems. The study also develops multi-objective evolutionary optimization methods for product family design problems. Multi-level commonality is considered for product family design with only parametric decisions or with both structural and parametric decisions. In each case, the commonality degree of the family is calculated to assess how same the product variants are to each other. A corresponding commonality control mechanism is also developed in the algorithm to determine which structural or parametric decision is common among which product variants in the family so as to achieve a trade-off between commonality and product performance. Computational results have demonstrated advantage