Thesis Opportunity: AI-Driven Monitoring and Decision Support for Automated Electrical End-of-Line Testing

Systemair is a leading company in ventilation with operations in 54 countries in Europe, North America, South America, the Middle East, Asia, Australia and Africa. Since the founding of Systemair in 1974, the company has shown positive operating profit. Over the past 10 years, average growth has reached about 10 percent. The company manufactures and markets energy-efficient and sustainable products that contribute to an improved indoor climate and reduced carbon dioxide emissions. We make sure that there is clean air that allows us humans to live and work together.

Are you passionate about AI and data analytics? In this thesis project, you will explore how AI can transform automated electrical end-of-line testing into a smarter and more predictive process. Background

Modern automated electrical end-of-line (EOL) testing systems generate large amounts of production and test data, including cycle times, test results, measurement values, failure codes, and process parameters. While this data is typically stored for traceability and quality assurance, its full potential is rarely utilized. Advances in Artificial Intelligence (AI) and Machine Learning (ML) provide opportunities to transform this data into actionable insights that improve productivity, quality, and process stability.

Objective

The objective of this thesis is to develop and evaluate an AI-based solution for monitoring and analyzing data collected during automated electrical end-of-line testing. The system should support data-driven decision-making by identifying trends, detecting anomalies, predicting process disturbances, and recommending process improvements.

Research Questions

The thesis can address all or selected parts of the following research questions:

• Can AI detect abnormal changes in cycle times and production throughput before they affect overall performance?
• Can machine learning identify patterns that precede increased failure rates or quality issues?
• Which process parameters have the greatest impact on test outcomes and product quality?
• Can AI-generated recommendations support optimization of test limits, setpoints, and process parameters?

Methodology

The project will utilize historical and real-time data from automated test systems, including:

• Cycle times and throughput data
• Test measurements and results
• Failure codes and defect statistics
• Product variants and production information
• Process and equipment parameters

Expected Outcomes

The project is expected to demonstrate how AI can:

• Detect process deviations and emerging quality issues at an early stage
• Monitor trends in cycle times, throughput, and failure rates
• Predict future process or quality problems
• Provide recommendations for optimizing test parameters and process settings
• Support continuous improvement initiatives through data-driven insights

Industrial Relevance

The results will contribute to the development of smarter manufacturing systems by leveraging existing production data to improve efficiency, reduce quality costs, and enhance process stability. The solution may serve as a foundation for future AI-assisted quality control and predictive process optimization in industrial production environments.

The student will gain hands-on experience in industrial AI applications, working with real manufacturing data and modern machine learning techniques in a production environment.

Additional information

The thesis project is connected to our operations in Skinnskatteberg. While some activities, meetings, and collaboration are best conducted on site, much of the work can be carried out remotely. We offer flexibility and will plan the level of on-site presence together based on the project's needs.

Submit your application no later than 2026-10-31. Applications will be reviewed on an ongoing basis, so we encourage you to apply as soon as possible.