In the project ‘Lane Analysis and Route Advisor’, we will conduct research into solutions based on advanced data analytics that combine the integration of various data sources (‘big data’), AI methods such as machine learning (ML), natural language processing (NLP), semantic web techniques, and optimization techniques for prescriptive analytics and decision making. These methods will be applied to the optimization of route planning in global freight forwarding, initially targeted at air freight shipment with special handling needs. The project that forms the context for this position also involves intensive collaboration with several industrial partners and academic partners (CWI and VU). The candidate will work at the School of Industrial Engineering, TU Eindhoven. The research project involves the Information Systems group (IS) and the Operations, Planning, Accounting and Control group (OPAC).
We are looking for a researcher who should have completed (or be close to completion of) a PhD degree in Computer Science, Artificial Intelligence, Econometrics, or Industrial Engineering, with a solid background in quantitative research methods. An ideal candidate should have a good basis in machine learning, or natural language processing, or semantic web technologies, and is interested in applications in logistics.
Privacy preservation becomes more and more important. However, what is the trade-off between privacy and decision-making quality? We conducted the first study on this trade-off that occurs when a k-anonymized database is used as input to the bin-packing optimization problem. Have a look at our paper if you are interested in this topic!
Very proud that our PhD student, Reza Refaei Afshar, has received the Best Student Paper Award for his work presented at the 11th International Conference on Agents and Artificial Intelligence held on 19 – 21 February, in Prague, Czech Republic. The paper titled “A Reinforcement Learning Method to Select Ad Networks in Waterfall Strategy” was co-authored by me, Murat Firat and Uzay Kaymak.
Nowadays, one of the most important sources of income for publishers who own websites is through online advertising (online ADs). For online publishers, it is difficult to design good strategies to manage their online AD auction systems due to highly dynamic real-time bidding environment. This paper proposes a machine learning based decision support system for publishers, which is built from historical AD auction data. The proposed method demonstrates its effectiveness in terms of the increased expected revenue for publishers.
The paper is available at: https://research.tue.nl/en/publications/a-reinforcement-learning-method-to-select-ad-networks-in-waterfal
PThe PhD position is part of the NWO funded project “MARCONI: Maritime Remote Control Tower for Service Logistics Innovation.” In this project, we aim to develop and demonstrate innovative service logistics concepts that exploit actual data on the state of maritime assets and the availability of the relevant maintenance resources. These concepts are aimed at (1) reducing maintenance costs, (2) increasing safety, by lowering the probability of unplanned system downtime and (3) reducing the number of unnecessary sailing movements (emissions) through smarter planning and/or clustering of maintenance activities. The ambition is to demonstrate the actual functioning of a remote service logistic control tower, with the long-term goal of developing and exploiting a scalable supply chain function in the maritime world. The PhD student will be a part of the research work-package on ‘Developing Service Logistics Decision Models’ led by TU/e. In the PhD project, there will be a close collaboration with the other partners of the MARCONI project: Boskalis, Damen, Gordian, Maastricht University, NLDA, Thales, RH Marine, Royal Netherlands Navy, and University of Twente.
See more information and apply here before March 10 2019: https://jobs.tue.nl/en/vacancy/phd-on-datadriven-maintenance-and-service-logistics-for-maritime-assets-455559.html
Super excited and proud! Our paper “Learning optimal classification trees using a binary linear program formulation” was accepted by AAAI (Thirty-Third AAAI Conference on Artificial Intelligence)! Acceptance rate 16.2% over 7700 submissions!
We show the power of optimization methods for deriving interpretable machine learning models.
Very happy that my paper “Fair task allocation in transportation”, co-authored with Charlie Ye and Rommert Dekker, was selected by Omega as “Best Paper Award for 2017”!
In this paper, we point out that in many cases, such as those in Sharing Economy, when allocating tasks and resources, cost should not be the major consideration. Distributing tasks/resources in a fair way among players is a more socially desired outcome. For achieving such outcomes, we design an efficient algorithm. We show by experiments that fairness often comes with a very small price in terms of cost. Check the paper out if you are interested!