Thesis Worker 30 hp - Integrating Wear and Sensor information

30 hp – Integrating Wear and Sensor information to predict Remaining Useful Life using Data-driven methods


Introduction

Thesis work is an excellent way to get closer to Scania and build relationships for the future. Many of today's employees began their Scania career with their degree project.

Background

TRATON GROUP is one of the world’s leading commercial vehicle manufacturers, with brands including Scania, MAN, International and Volkswagen Truck & Bus. Its portfolio covers light commercial vehicles, trucks and buses, complemented by financing, charging and digital logistics services. Through its global operations, production sites and extensive sales and service networks, TRATON has access to diverse vehicle platforms, real-world operational data and fleet deployment environments. This provides a strong foundation for developing and validating innovative solutions for sustainable and efficient transportation. 

In the Cloud and Embedded Plattform domain within TRATON, we develop new solutions for connected vehicles in our Internet of Things (IoT) platform, as part of TRATON’s increasing focus on communication, services and smart transport solutions. Advanced data analysis capabilities are a cornerstone enabler in this development.  

Target/scope 

Modern connected vehicles and industrial systems are increasingly equipped with sensors that continuously monitor their condition. This enables a shift from reactive to predictive maintenance, where decisions are based on the expected Remaining Useful Life (RUL) of components. 

RUL prediction relies on two complementary information sources: wear factors, which describe accumulated degradation through usage and operating conditions, and sensor factors, which reflect the current observed state of the system. 

Our recent work [1] proposes a probabilistic framework that combines these two sources using a survival-based aging model and a data-driven model, enabling not only RUL estimation but also associated prediction uncertainty. 

The objective of this thesis is to implement and extend this framework, and to investigate how different modelling and uncertainty quantification choices affect RUL prediction performance and reliability. 

Job description

The thesis will investigate probabilistic fusion of wear-based and sensor-based models for Remaining Useful Life prediction. This work will include: 

1. Reproduce and validate the proposed framework: Implement and evaluate the approach presented in [1], including the survival-based aging model, neural-network-based RUL model, and their probabilistic combination. 

2. Explore alternative modelling approaches: Investigate alternatives to the Weibull survival model and/or neural-network architecture used in [1]. Evaluate how different modelling assumptions influence RUL prediction and uncertainty. 

3. Evaluate generalisation across datasets: Test the framework on additional datasets and operating conditions to study its robustness and ability to generalise beyond the original experimental setting. 

4. Investigate uncertainty quantification for RUL prediction: Compare different approaches for estimating predictive uncertainty and study their effect on the resulting RUL distributions, prediction intervals, calibration and accuracy. 

If time permits, investigate extensions that relax the proxy assumption used in the data-generating formulation in [1], with the aim of developing a more general relationship between wear, latent component health, sensor observations, and failure. 

The thesis provides an opportunity to work at the intersection of machine learning, survival analysis, uncertainty quantification and predictive maintenance, with both methodological research questions and applications to real-world industrial systems. 

Reference

[1] Srinivasan, Abhishek, et al. "Integrating Survival-Based Aging Models with Data-Driven RUL Prognostics." PHM Society European Conference. Vol. 9. No. 1. 2026. 

Education/line/direction 

Masters programmes in Machine Learning, Data Science, Computer Science, Complex Adaptive Systems or similar.  

Number of students: 1 

Start date for the Thesis project: Spring 2027 

Estimated timescale: 20 weeks 

Contact person and supervisor 

Abhishek Srinivasan, Data Scientist, 08-553 816 96, abhishek.srinivasan@scania.com 

Juan Carlos Andresen,Group Manager, 08-553 835 16, juan-carlos.andresen@scania.com 

Application

Your application should contain the following:  

• CV, 

• personal letter, 

• and copies of grades.
  
  
  

Until 2026-10-31. Applicants will be assessed on a continuous basis until the position is filled.