As a Graduate Intern at ORTEC you will be part of a unique organization and team!
The Train Unit Shunting Problem (TUSP) is a difficult sequential decision-making problem faced by Dutch Railways (NS). At a service site, several train units need to be inspected, cleaned, repaired and parked during the night. Also, specific type of train units need to depart at a specific times. All these requirements and constraints make the TUSP a very challenging scheduling and planning problem.
NS has started to explore the possibility to solve the TUSP problem with Deep Reinforcement Learning, and this has shown great potential in solving the parking problem using Deep Q-learning with Convolutional Neural Networks.
Recent research from ORTEC has shown encouraging results for solving routing problems using the REINFORCE algorithm and a model based on Neural Attention. The goal of this project is to join forces between ORTEC and NS and develop a strategy based on this approach for the TUSP. This way, we hope to be more capable of dealing with uncertainty and delivering significantly more consistent solutions compared to approaches currently being developed by NS.
This is a joint project between NS and ORTEC, and the master student is expected to spend 50% of time at NS (Utrecht) and 50% of time at ORTEC.
If you are interested, we would like to receive your resume, motivation and grade list. You can send your application to firstname.lastname@example.org. If you want to know more about the possibilities or need more information, then please contact Gordon Boon (Recruiter).
Who you are
What we offer
What to expect
We help you to thrive in your field of expertise. We operate a flat organizational structure that keeps communication lines short. The atmosphere is open, informal, cooperative and positive. We employ over 900 people in the Netherlands (HQ), Belgium, Germany, France, the U.K., Romania, Italy, the U.S., Australia, Brazil, Poland and Denmark.
Visit our website www.ortec.com to learn more about our solutions and clients’ experiences.
Acquisition as a result of this vacancy is not appreciated.