Enhanced Capability Monitoring of Production Lines for Philips Personal Health Products

Varoščić, Nika (2026) Enhanced Capability Monitoring of Production Lines for Philips Personal Health Products. Internship report thesis, Data Science and Society (DSS).

Text Final Internship Report - Nika Varoščić (S5705606).pdf Restricted to Registered users only Download (5MB)

Abstract

During my studies in the BSc Data Science and Society programme, I have gained experience in data analysis, statistical reasoning, and the application of machine learning techniques to both societal and technical problems. Throughout the programme, I developed a strong interest in machine learning as a tool for extracting structure and insight from complex data. This was reflected in several academic projects, including a field project on the synthesis of municipal housing data using agent-based modelling to enable privacy-preserving data sharing, as well as a simulation exercise in which I worked as part of an AI and retrieval-augmented generation (RAG) team on a context- aware learning assistant. While these projects were not focused on forecasting, they strengthened my motivation to work with real-world data and to further explore how data-driven models can support decision-making in practice. Alongside this, I became interested about predictive questions more broadly, particularly how historical data can be used to anticipate future behaviour. This motivation led me to pursue an internship at Philips Drachten, which started in September 2025 and lasted until the end of December 2025. I was particularly drawn to the opportunity to gain hands-on experience with real production data within a globally recognised company. Working at Philips offered the chance to move beyond controlled academic datasets and engage with complex, noisy, and operational data, providing a valuable learning experience. This report reflects that learning journey and documents the exploration and comparison of multiple forecasting approaches. The work presented here would not have been possible without the guidance and support of my supervisors. I would like to thank Dr. Noman Haleem, my university supervisor, for his academic guidance and feedback, as well as Ing. André Stefan, my supervisor at Philips, for his support, domain expertise, and constructive discussions throughout the internship. Overall, this internship provided a valuable opportunity to bridge theory and practice and to deepen my understanding of machine learning and forecasting within real-world industrial applications.

Item Type:	Thesis (Internship report)
Name supervisor:	Haleem, Noman
Date Deposited:	09 Feb 2026 13:27
Last Modified:	09 Feb 2026 13:27
URI:	https://campus-fryslan.studenttheses.ub.rug.nl/id/eprint/782

Actions (login required)

View Item