Developing an Effective Predictive Model for Imminent Component X Failures in Heavy-Duty Scania Trucks
We are happy to announce the industrial challenge supported by SCANIA! This page contains the summarized information about the challenge. Thus, the participants are strongly encouraged to carefully read the challenge call to comply with the requirements and instructions.
The goal is to develop a predictive model (or models) that can accurately predict whether a specific engine component (hereinafter referred to as ‘Component X‘) in a vehicle is at risk of imminent failure. The dataset includes information gathered from Scania trucks used in heavy-duty applications.
You should send your submission to our industrial challenge chair, Tony Lindgren (firstname.lastname@example.org). The email should have the title ‘IDA 2024 Industrial Challenge Submission’. The submission requires the following files:
- IDA_Industrial_challenge_2024_predictions.csv: The prediction results from your prediction model.
- IDA_Industrial_challenge_2024_paper.pdf: A paper that explains your prediction model.
The paper should be 6-8 pages and follow the format of regular paper submissions using the Lecture Notes of Computer Science (LNCS) format. Submissions that do not follow any of the requirements may not be considered. Submission details (e.g., the contents of the prediction result file) can be found in the challenge call. Please read the instructions carefully!
The top three contestants will be included in the IDA conference proceedings and they will also present their methodology and results at the conference in the industrial challenge session. Scania has generously provided prize money to the top three contenders, where the first-placed contender/team will receive an amount of 5000 SEK, the second-place contender/team will receive an amount of 3000 SEK, and the third-place contender/team will receive an amount of 2000 SEK.
The datasets can be accessed here. The data for the challenge comprises of multiple files, each containing different types of information. Where the operational training-, validation- and test-data have identical information content layout. Please make sure that you have downloaded all datasets before you start to work on the challenge.
More information, such as the description of the data, can be found in the challenge call. The participants are also encouraged to read the SCANIA Component X Dataset paper, which guides through the dataset structure, and cite this paper in the submission paper when referring to the dataset.