Trade-off Management in Early Mechanical Design: Identifying, Understanding, and Mitigating the Causes of Trade‐offs Between Design Objectives

Trade‐off situations are inescapable in product development. No product can be infinitely durable, sustainable, inexpensive, efficient, user friendly, and so on. Many of these tradeoffs occur due to challenges within the domain of mechanical engineering. Make a structure lighter, and it usually loses stiffness. Make a mechanism more accurate, and it usually becomes less mechanically efficient. These situations often arise due to inherent dependencies between the goals a product is designed towards. Yet, they can also emerge due to the constraints that follow the manufacturing processes and materials used to manufacture these products.

This challenge will only grow over time. Competitive pressures drive companies to strive to improve product performance and integrate more features, all the while keeping costs low. In turn, this drives designers to attempt to realise as much functionality with as few components as possible, with more trade‐offs arising as a result. This tendency will only increase in the future with societal needs and technological developments introducing more and more objectives that the design engineer needs to take into account. An example of this is the drastically increasing need to develop more sustainable products.

While compromise may at times be inevitable, the lack of up‐front awareness, understanding, and mitigation of trade‐offs during the initial stages of mechanical design can have substantial consequences for the performance of the end product. This can also delay product development projects and result in unforeseen quality and cost issues in production. In other words, trade‐offs can delay technological progress in general.

This PhD thesis describes the development of methods for the management of trade‐off situations during the early phases of mechanical design. This includes Pareto set Dependency Analysis, a quantitatively founded approach that that builds upon existing monotonicity analysis and design optimisation methods to identify trade‐offs and their underlying root causes. This opportunistic yet rigorous approach led to the development redesign and synthesis methods that allow the identification of design changes that result in an improved Pareto set. This implies that the trade‐offs have been mitigated or reduced and that an overall improvement in product performance has been achieved. This provides a systematic foundation allowing designers to continually identify design changes that result in improved performance, as the design of the product is gradually refined, even if the need for additional features, functionalities, and requirements arises.

This research was conducted in an industrial‐academic collaboration between DTU Mechanical Engineering and Novo Nordisk. Cases from ongoing product development projects were used in the research, one of which is included in this thesis. The SOMA device, an ingestible medical device for the oral delivery of pharmaceutical compounds such as insulin, is used to demonstrate the application of these analysis and redesign methods. Using the novel analysis methods and multiobjective optimisation, several drivers of trade‐offs in the SOMA device were identified. Many of these were successfully mitigated using the redesign methodology, resulting in a set of redesigns that exhibit improved Pareto sets, confirming the practical value of the results of the research.