Functional modelling for design concept optimisation (FuMO)
PhD student: Mostafa Meliani
Started: February 2019
The project is focussed on functional modelling and the trade-offs between accuracy and simplicity in building, analyzing and interpreting the results. Typically, the more detailed the model, the more time is needed to evaluate it, and there is a clear conflict between the computational complexity and the accuracy of a model. This project aims to study structured methods and criteria to select suitable models for a particular problem. To search for an economical as well as ecological solution to a design problem, optimization can be used. The first step of any optimization is to create a model of the system to be optimized and the success of an optimization is very much dependent on the model used. Optimization in itself is about finding trade-off solutions to cross-functional conflicts. In this case, using suitable models for the different functions might be even more critical, since models in different disciplines typically have highly varying computational complexity and precision and it is key to have a good balance between the models to use the computational budget in the best possible way. This PhD project is also a part of a co-tutuelle agreement with University of Eastern Finland, contributing with their expertise on inverse modelling and Bayesian methods.