Skip to main content
  • CIS
    Members: Free
    IEEE Members: Free
    Non-members: Free
    Length: 00:49:14
23 Jul 2020

Evolutionary optimization applied to engineering design has been very successful in the last two decades. While there are still many challenges, the recent advances in machine learning and data analysis have intensified the interest in expanding the role of optimization and AI in the engineering design process conceptually. Three areas have been in the focus of the work in our group, which I will discuss in my presentation. First, using the term cooperative intelligence, we emphasize new ways that engineers and optimization systems can interact, work and grow together. Complementary skill sets such as processing large data, automatic model building or complex system simulations enrich human intuition and design thinking for increased overall system performance. Secondly, I will report on experienced-based computation, where we aim at capturing abstracted knowledge about one optimization process in a way that allows it to be applied to a different albeit related task. Therefore, after each optimization the systems should gather more experience and become more knowledgeable about the problem class allowing for higher-level abstractions and rapid reconfigurations – in the same way as a senior engineer has amassed experience over time that is different from textbook knowledge and that enables her to judge solutions differently as compared to a junior researcher. In my presentation, I will focus on the role of representations in combination with learning for capturing valuable experience in design optimization.
Thirdly, I will discuss the application of optimization to finding conceptually novel solutions to a problem. Novelty or interestingness of solutions need to be defined in relation to the objectives and the already explored part of the design space. A creative engineering process leading to disruptive innovations in technology is mostly the result of a team effort including the ability to cope with dynamic environments. The combination of novelty optimization and cooperative intelligence will enable the AI system to be part of this team.
Cooperation, Experience and Creativity together are key enablers for the next generation of optimization and AI systems to provide qualitatively new value to the engineering process.