Date of Award

7-28-2015

Publication Type

Doctoral Thesis

Degree Name

Ph.D.

Department

Industrial and Manufacturing Systems Engineering

Keywords

Assembly Sequence Planning, Association Rules, Data Mining, Knowledge Discovery, Manufacturing System Synthesis, Product Design Retrieval

Supervisor

ElMaraghy, Hoda

Rights

info:eu-repo/semantics/openAccess

Abstract

The variety of products has been growing over the last few decades so that the challenges for designers and manufacturers to enhance their design and manufacturing capabilities, responsively and cost-effectively are greater than ever. The main objective of this research is to help designers and manufacturers cope with the increasing variety management challenges by exploiting the data records of existing or old products, along with appropriate Knowledge Discovery (KD) models, in order to extract the embedded knowledge in such data and use it to speed-up the development of new products. Four product development activities have been successfully addressed in this research: product design, product family formation, assembly sequencing and manufacturing system synthesis. The models and methods developed in this dissertation present a package of knowledge-based solutions that can greatly support product designers and manufacturers at various stages of the product development and manufacturing planning stages. For design retrieval; using efficient tree reconciliation algorithms found in Biological Sciences, a novel Bill of Materials (BOM) trees matching method was developed to retrieve the closest old design and discover components and structure shared with new product design. As a further application to BOM matching, an enhanced BOM matching method was also developed and used for product family formation. A new approach was introduced for assembly sequencing, based on the notion of consensus trees used in evolutionary studies, to overcome the critical limitation of individual assembly sequence retrieval methods that are not able to capture the assembly sequence data for a given new combination of components that never existed before in the same product variant. For manufacturing system synthesis; a novel Integer Programming model was developed to extract association rules between the product design domain and manufacturing domain to be used for synthesizing a manufacturing/assembly system for new products. Examples of real products were used to demonstrate and validate the developed models and comparisons with related existing methods were carried out to demonstrate the advantages of the developed models. The outcomes of this research provide efficient, and easy to implement knowledge-based solutions for facilitating cost-effective and rapid product development activities.

Share

COinS