AutoFix Automated Design of fixtures
| Reference number | |
| Coordinator | Linköpings universitet - Avdelningen för industriell ekonomi/IEI |
| Funding from Vinnova | SEK 4 240 682 |
| Project duration | October 2020 - December 2023 |
| Status | Completed |
| Venture | FFI - Sustainable Production |
| Call | Sustainable production - FFI - June 2020 |
| End-of-project report | 2020-02974svenska.pdf(pdf, 2093 kB) (In Swedish) |
Important results from the project
The purpose of the AutoFix project was to increase the degree of automation in fixture design based on multidisciplinary optimization (MDO) and machine learning (ML). AutoFix has developed methods and tools to automatically optimize resource-intensive fixture work using design automation (DA), MDO, and ML. All work packages have been completed, and from a technical perspective, knowledge generation has been significant, especially in machine learning, particularly in areas such as reinforcement learning and supervised learning.
Expected long term effects
The project resulted in four scientific publications and a licentiate degree, fully or partially financed by AutoFix. The project has resulted in a significant increase in knowledge increase within academia and the technology continues to be implemented and developed in other research projects.
Approach and implementation
All work packages have been completed, and from a technical perspective the knowledge generation has been significant, especially in machine learning and in areas such as reinforcement learning and supervised learning. Despite this, the project did not achieve its market objective. Models and frameworks have been developed in tools that Volvo does not use and have therefore been difficult to implement on site The implementations have not solved a big enough problem with respect to investments needed to develop them.