The goal of the project has been to improve the tyre selection at Volvo Group Trucks Technology, through developing a methodology using mathematical optimization. Volvo wants to give recommendations to their truck users of which tyres they should select, in order to save fuel. This is done when the tractive energy required to run the vehicle is reduced. Which tyres are optimal depends on many aspects, such as the environment where the truck will be used, how the load will be distributed, and the model of the truck. This create lots of optimization problems, that needs to be solved almost instantly as the customers should not have to wait for an answer.
Zuzana has developed a splitting algorithm for a large number of instances of simulation-based optimization problem with categorical variables of no natural order. A database of strategic vehicle specifications and their usage of the vehicles has been set. In that way a lot of the computation is done beforehand, and when a new customer arrives it is possible to identify the optimal tyres using the solutions in the database.
When the algorithm was tested, the results showed that up to 3% of the tractive energy can be saved with the optimal tyres selected when compared to the tyres that are sold with the truck. The methodology can however not be applied immediately. There is a new project on its way to propagate tyre data, the vehicle models need to be developed automatically, and more data is also needed from the tyre suppliers. When these parts are solved, the project is ready to be implemented.
The project was initiated by Volvo Group Trucks Technology and is also funded by the Swedish Energy Agency. The optimization group at the Department of Mathematical Sciences has cooperated with the Department of Mechanical and Maritime Sciences, to develop the vehicle model, the Institute of Technology at the University of Ontario, to develop the tyre model, and Fraunhofer-Chalmers Research Centre Industrial Mathematics, to develop the algorithm models.
Zuzana took a master course in optimization at Chalmers in 2010 through the Erasmus programme, and found that she liked the subject a lot. The PhD project she has now completed was announced at the time she finished her master studies, she took a chance, and so it happened that she stayed in Sweden six more years than originally intended. After her dissertation she will however move back to the Czech Republic, where she will work with optimization of electric grids for an electric power distribution company.
– This has, overall, been a good experience. A very positive part of it is the Swedish parental leave system, allowing to combine career and family very easily.
Zuzana Nedělková will defend her PhD thesis “Optimization of truck tyres selection” on February 22 at 10.15 in the room Pascal, Hörsalsvägen 1. Supervisors are Ann-Brith Strömberg and Peter Lindroth, co-supervisors are Michael Patriksson and Bengt Jacobson.
Text and photo: Setta Aspström
Illustration: Zuzana Nedělková (cover of the PhD thesis)