PREST:Predictive Strategy using Machine Learning for Smart Test Case Selection



Start date:


End date:


The objective of the PREST project is to explore a smart and predictive strategy using Artificial Intelligence and Machine Learning for test case selection. The proposed predictive strategy can help to achieve various goals including test prioritization, handle flakiness phenomenon, detect early regression test in software performance, and provide a recommendation to remove test cases that are obsolescent. PREST will give us the knowledge needed to propose a new machine learning approach for smart test case selection which will be covered by current literature, standards and practices. Based on these findings we plan to develop a benchmarking methodology as a proof-of-concept in a systematic way such that the industrial partners be able to perform better and more efficient software testing that could improve the quality of the product and business value. The project relies on relevant input data and information provided by the project partners Volvo Car Corporation and Volvo Technology AB (Volvo Trucks). Their systems will be used to extract relevant use-cases needed in this project. We expect them to contribute to this project by providing an industrial perspective as that is required for the success of this project.

First NameLast NameTitle
Shaibal Barua Senior Lecturer
Volvo Cars Industrial
Volvo Group Trucks Technology Industrial

Shaibal Barua, Senior Lecturer

Room: U1-050A