Platform Design for Producibility: Early-Stage Modeling and Assessment Support
Doctoral thesis, 2018
In industry, platforms are commonly adopted to reduce unique parts among a variety of distinct product variants, which have proven to be cost-effective within a single platform lifecycle. However, when the platform becomes obsolete or modifications are required to capture changing customer and production needs and requirements, manufacturers often spill tears over the time-consuming and costly processes of reusing and adapting the current platform structure into new. In design, such a platform structure of parts is rigid and often characterized by redundant data and weak relations among and across product variants and existing production machinery. To improve the ability to reuse design and production information for assessing new concepts more quickly, non-rigid platform representations of product concepts and existing production machinery are necessary but not clarified in literature and rarely implemented in industry. In this thesis, research studies have therefore been conducted to (1) investigate how early-stage information about a variety of products and existing production machinery can be represented to improve design-production responsiveness, and (2) develop methods and tools to model and generate a set of product-production alternatives as a basis for producibility assessments. A number of engineering case studies have been prepared by researchers and industrial specialists. Data, related to product and production variety and their mutual constraining factors, have been collected by interviewing industrial specialists, as well as examining corporate documents of both product design prerequisites and capabilities in production. The engineering case studies prepared have supported the creation of new knowledge and been used to demonstrate the usefulness of the improved models, methods and tool devised supporting platform design for producibility. As opposed to rigid parts, findings show that platform entities can be represented as reusable and adaptable system objects containing early-stage information of product variety, existing production resources and processes. This information mainly consists of a common product-production structure of relations among functional requirements, design solutions, mutual constraining factors and target values. By creating a complementary producibility system, including rule-based and simulation- based models, early-stage producibility assessments of product concepts can be supported. Findings emphasize the dynamic consideration of producibility during the platform design as customer and production needs and requirements frequently change. By employing the early-stage modeling and assessment support devised, manufacturers can (1) represent product and production variety as reusable and adaptable system objects with links to producibility constraints, available over generations of products and production systems and (2) dynamically and concurrently model, generate and assess product-production alternatives under producibility constraints during early design stages as a basis for putting inferior alternatives aside until new information becomes available. Theoretically, the number of costly and time-delaying late- stage modifications of product designs, production configurations or both can be reduced. However, to validate and generalize these hypothetical effects, they need to be measured in future studies.
Reuse of design and production information
Enhanced function-means modeling
Set-Based Concurrent Engineering