Bachelor and Master Theses

Background

In heavy machinery maintenance, particularly undercarriage inspection, effective decision-making requires integrating multiple sources of information, including visual inspection results, component wear measurements, operating conditions, machinery age, and historical maintenance records. Conventional machine learning approaches typically focus on individual tasks, such as component detection, wear estimation, or failure prediction. However, transforming these outputs into context-aware maintenance recommendations often requires additional decision-making processes and domain expertise.

Recent advances in Agentic Artificial Intelligence (Agentic AI) offer new opportunities to develop intelligent systems that can autonomously coordinate multiple analytical tasks, retrieve relevant information, reason over contextual data, and recommend appropriate actions. An agentic system could interpret wear measurements obtained from image segmentation, retrieve historical maintenance information, assess component conditions, and recommend whether a component requires further inspection, maintenance, or replacement. Such recommendations could consider maintenance costs, operational constraints, component lifetime, and sustainability-related factors.

The aim of this thesis is to develop and evaluate an agentic AI framework for prescriptive analytics that integrates object segmentation outputs, contextual information, and domain knowledge to generate explainable and actionable maintenance recommendations for heavy machinery undercarriage components.

Thesis Objectives

The main objective is to design, implement, and evaluate an agentic, AI-based prescriptive analytics framework that integrates visual inspection results and contextual information to support maintenance decision-making.

The specific objectives are to:

1.    Develop an agentic framework that integrates object segmentation and measurement outputs with machinery operating conditions, maintenance history, and domain knowledge.

2.    Implement intelligent agents capable of information retrieval, condition assessment, and maintenance recommendation generation.

3.    Evaluate the framework in terms of recommendation quality, decision consistency, explainability, computational efficiency, and adaptability to different machinery configurations.

Expected Outcomes

A proof-of-concept agentic AI framework that integrates visual inspection outputs and contextual information to enable prescriptive maintenance.

An agent-based decision-making pipeline that coordinates information retrieval, condition assessment, and maintenance recommendation generation.

Context-aware and explainable recommendations for undercarriage inspection, maintenance, and component replacement. 

An empirical evaluation of the proposed framework against conventional rule-based or non-agentic decision-making approaches.

• Agentic AI & LLM Frameworks: Understanding multi-agent architectures, tool use, reasoning loops (e.g., ReAct), planning, and context management using Large Language Models (LLMs) or smaller specialized models.

• Maintenance Engineering & Prescriptive Analytics: Familiarity with maintenance strategies (preventive, predictive, prescriptive), Remaining Useful Life (RUL) estimation, and cost-benefit optimization in industrial settings.

• Explainable AI (XAI): Principles of making AI decisions transparent, traceable, and understandable for human operators and domain experts.

• Programming & Framework Proficiency: Advanced proficiency in Python and standard deep learning frameworks (PyTorch or TensorFlow), alongside experience with open-source generative libraries (e.g., Hugging Face diffusers, torchvision).

Volvo Construction Equipment (VCE), a division of Volvo Group, is one of the world's largest manufacturers of construction equipment such as wheel loaders, haulers, excavators and compact machines.

At VCE, we are at the forefront of innovation in the construction equipment industry. We focus on data-driven decision-making across the entire business spectrum. With numerous ongoing exploratory projects, we provide a fertile ground for curious minds to delve into diverse domains and leave a lasting impact.

We seek individuals with innovative mindsets who can provide unexpected solutions and think outside the box. By joining us, you will contribute to research and findings that will help shape the direction of our business. Together, we will explore untapped potential, paving the way for a more efficient, intelligent, and sustainable future.