WO Stage - PQI Slab Logistics

An average day…

You will work in the Product Quality & Innovation (PQI) Department, together with researchers from different technical backgrounds. Your work will focus on improving the use of hot rooms in slab logistics at Tata Steel Nederland. The goal is to support better allocation decisions that keep slabs hot for longer, while avoiding unnecessary crane operations and reshuffling.

You will build on an existing slab logistics model and previously developed heuristic approaches. Your main task will be to investigate how AI methods can support smarter hot-room allocation and help evaluate future logistics concepts.

What are you going to do?

Steel slabs lose thermal energy during storage and transport between production areas and the hot strip mill. Hot rooms can reduce this heat loss, but their value depends strongly on how they are used. Poor allocation may lead to unnecessary slab movements, extra crane operations, and limited operational acceptance.

In this internship, you will develop and evaluate an AI-based decision-support approach for hot-room allocation. The preferred direction is a hybrid AI method, where existing heuristic or meta-heuristic solutions are used to train an initial decision model, which may later be improved using reinforcement learning or related techniques.

The model should help answer questions such as:

• how to increase hot-room utilization without excessive crane movements;
• whether additional or mobile hot rooms provide operational value;
• whether heated hot rooms could further improve energy efficiency.

The work will be performed using simulation scenarios and available process/logistics data where possible. The result should be a compact prototype and evaluation showing how AI can support future energy-efficient slab logistics.

This assignment is intended for an MSc student in Computer Science, Artificial Intelligence, Data Science, Operations Research, or a related field. Furthermore, you have:

• strong Python programming skills;
• knowledge of machine learning;
• interest in optimization, logistics, or simulation;
• ability to work independently and structure a technical problem clearly.

Experience with reinforcement learning, imitation learning, PyTorch, simulation models, or industrial scheduling is considered an advantage.

• A 3- to 4-month MSc internship assignment.
• A monthly internship allowance.
• A challenging project at the interface of AI, logistics, and energy efficiency.
• Supervision from researchers and domain experts.
• The opportunity to contribute to Tata Steel Nederland’s transition towards smarter and more sustainable steel production.

What is your working environment?

You will work in an informal, research-oriented environment where practical impact and technical quality are both important. You will interact with researchers and operational experts to understand the logistics context and evaluate whether the developed AI approach is useful for real industrial decision-making. The internship will be carried out mainly at our office, with the required equipment and support provided.

Tata Steel challenges you!

Tata Steel Nederland is working on major changes in steel production and energy use. Smarter logistics and better heat retention can contribute to reducing reheating demand, natural gas consumption, and CO2 emissions. Are you interested in applying AI to a real industrial sustainability challenge? Then this internship offers the opportunity to work on a concrete problem with both scientific and operational relevance.

• nieuwsgierige instelling
• student aan wo opleiding
• zelfstandigheid