A GENERATIVE MODELING APPROACH TO ENGINEERING DESIGN
Ola Isaksson
Abstract
Ola Isaksson
Abstract
This paper describes an approach at Volvo Aero to systematically develop and use design and evaluation support systems, using Knowledge Based Engineering (KBE) generative modeling techniques. Product and process knowledge derived from product strategies and experiences from previous design projects have to be organised into a successful generative model. These generative models support both the synthesis stages of design and the analysis stages, allowing design alternatives to be compared with each other. The designs are described in product models with a full 3D solid and shell representation, together with material and manufacturing information. The organisation of these models uses ideas from well-known theories in engineering design to meet the requirement of flexibility in updating and reorganisation of the generative models. An example of how these generative design and evaluation systems have been used for a jet engine component conceptual design study is presented. The lead-time for a design and evaluation loop was shortened by more than 90%, which allowed multiple candidates to be evaluated and presented at the design review following the design study. The design study also required updating of the design system, which was included in the lead time for design. Generative Modeling, Engineering Design, Aircraft Engine, Methodology, Knowledge Based Engineering
OpenAlex reports 21 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
This paper describes an approach at Volvo Aero to systematically develop and use design and evaluation support systems, using Knowledge Based Engineering (KBE) generative modeling techniques. Product and process knowledge derived from product strategies and experiences from previous design projects have to be organised into a successful generative model. These generative models support both the synthesis stages of design and the analysis stages, allowing design alternatives to be compared with each other. The designs are described in product models with a full 3D solid and shell representation, together with material and manufacturing information. The organisation of these models uses ideas from well-known theories in engineering design to meet the requirement of flexibility in updating and reorganisation of the generative models. An example of how these generative design and evaluation systems have been used for a jet engine component conceptual design study is presented. The lead-time for a design and evaluation loop was shortened by more than 90%, which allowed multiple candidates to be evaluated and presented at the design review following the design study. The design study also required updating of the design system, which was included in the lead time for design. Generative Modeling, Engineering Design, Aircraft Engine, Methodology, Knowledge Based Engineering
Key concepts: Generative Design, Engineering design process, Probabilistic design, Generative grammar, Systems engineering, Product design, Flexibility (engineering), Computer science