PhD in Explainability and Humans in the Loop in Optimization Processes

Introduction

Are you passionate about human-centric AI and explainable decision-making? Are you intrigued by the role of human expertise in complex optimization? This PhD position focuses on bringing humans into the loop of optimization processes, with the goal of improving decision quality and transparency through explainable AI. You’ll investigate how domain knowledge from human decision-makers can be formalized and effectively integrated into different stages of optimization, both as priors and as feedback after optimization outcomes.

Job Description

This PhD project is a part of the CoRDS project – Confident Data-Driven Decision Support (https://www.cords-dn.at). CoRDS is a large-scale Doctoral Network, funded by the European Union under the Marie Skłodowska-Curie Actions (MSCA) Doctoral Networks programme Grant Agreement No. 101227512. It brings together 8 European universities and 14 societal partners in an ambitious interdisciplinary collaboration. Our shared mission is to advance the research on data-driven optimization methods and train a new generation of experts skilled in the combination of Operation Research and trustworthy Machine Learning. CoRDS goes beyond the state-of-the-art DDO methods by proposing decision support frameworks that combine OR and ML methods and enable robust, transparent, and fair solutions that reflect user preferences and address complex, uncertain real-world scenarios.

This PhD position belongs to the work package WP6: Transparency. As part of this doctoral project, you will develop novel knowledge representation techniques, algorithms for human-in-the-loop optimization, and interactive tools that combine graphical and natural language interfaces to gather and incorporate expert knowledge. A central research question is how human mental models of an optimization problem align or conflict with what optimization algorithms need, and how explanations can mediate that gap across the full optimization cycle.

Objectives

(1) Characterize the types of domain knowledge that human decision-makers hold about optimization problems, and formalize a representation that is both computationally tractable and interpretable by non-technical experts.
(2) Develop a bidirectional HITL framework where explanations mediate knowledge injection before optimization and feedback integration after, with explicit models of when and how human input should be weighted against algorithmic outputs.
(3) Evaluate the framework on trust calibration, knowledge elicitation consistency, and solution quality across at least two optimization domains and two human roles, studying whether explanations help humans decide when to intervene versus when to defer to the algorithm.
(4) Investigate the generalization of the proposed methods across optimization problem classes, and characterize the conditions under which human-in-the-loop interaction improves or degrades solution quality.

Expected Results

(1) A typology of domain knowledge types relevant to optimization settings, accompanied by a knowledge representation formalism that is feasible to integrate with optimization algorithms at different stages of the solving process.
(2) A set of algorithms for integrating domain knowledge before optimization as priors and after optimization in the form of structured expert feedback, with theoretical and empirical analysis of the conditions under which each integration mode is most beneficial.
(3) A validation framework assessing the internal consistency of elicited knowledge, the fidelity and utility of generated explanations, and the trade-off between solution optimality and human acceptance across different problem settings and user profiles.
(4) A tool with an interface combining graphical and natural language interaction to gather domain constraints and expert feedback in a structured way, tested across at least two domains and validated with users from different roles.

This position will be based in the Information Systems (IS) Group at the Department of Industrial Engineering & Innovation Sciences, Eindhoven University of Technology (TU/e). The candidate will be supervised by Dr. Isel Grau and Dr. Yingqian Zhang (from TU/e). As part of the project you will also take part in two secondments, one at a industry partner, where you will test early prototypes in real logistic scenarios, and another at the University of Vienna, where you will collaborate with other PIs and PhDs of the project.

Note: Strong applications will be reviewed on a rolling basis. We encourage candidates to apply early and not wait for the deadline.

Job Requirements

• A Master's degree in Artificial Intelligence, Computer Science, Data Science, Operations Research, Industrial Engineering, or a closely related field.
• A motivated candidate with a research-driven mindset and a genuine interest in advancing human-centric AI for real-world decision support, in close collaboration with industry.
• Strong skills in machine learning, explainable AI, and programming, with experience in optimization methods or a strong motivation to develop it.
• Preferably with a background or demonstrated interest in human-computer interaction, cognitive aspects of decision-making, or user studies involving domain experts.
• Familiarity with or strong motivation to develop skills in experimental evaluation methods, including user studies, expert elicitation, and empirical assessment of AI systems.
• Experience with or strong interest in building conversational agents or dialogue interfaces, including working with large language models, prompt engineering, and natural language generation for interactive applications.
• Ability to work independently and collaboratively in an interdisciplinary team spanning operations research, machine learning, and human factors.
• Motivated to develop your teaching skills and coach students.
• Fluent in spoken and written English (C1 level).

Conditions of Employment

A meaningful job in a dynamic and ambitious university, in an interdisciplinary setting and within an international network. You will work on a beautiful, green campus within walking distance of the central train station. In addition, we offer you: 

• Full-time employment for four years, with an intermediate assessment after nine months. You will spend a minimum of 10% of your four-year employment on teaching tasks, with a maximum of 15% per year of your employment. 
• Salary and benefits (such as a pension scheme, paid pregnancy and maternity leave, partially paid parental leave) in accordance with the Collective Labour Agreement for Dutch Universities, scale P (min. € 3,059 - max. € 3,881 gross per month).  
• A year-end bonus of 8.3% and annual vacation pay of 8%. 
• High-quality training programs and other support to grow into a self-aware, autonomous scientific researcher. At TU/e we challenge you to take charge of your own learning process. 
• An excellent technical infrastructure, on-campus children's day care and sports facilities.  
• Unlimited access to the modern on‑campus TU/e Student Sports Center at an exceptionally affordable rate.
• An allowance for commuting, working from home and internet costs. 
• A Staff Immigration Team and a tax compensation scheme (the 30% facility) for international candidates. 

On our website you can discover even more information about our conditions of employment. Build on your career at TU/e!

About us

We are a leading international university where scientific curiosity meets a hands-on mindset. We work in an open and collaborative way with high-tech industries to tackle complex societal challenges. Our responsible and respectful approach ensures impact — today and in the future. TU/e is home to over 13,000 students and more than 7,000 staff, forming a diverse and vibrant academic community.

Our university is located in Brainport Eindhoven — a world‑leading tech region with more than 7,000 high‑tech companies and strong R&D activity. Known for breakthroughs in AI, photonics, semiconductors and advanced manufacturing, Brainport is a place where technology serves people and society. Learn more about the Brainport region here.

The Industrial Engineering & Innovation Sciences (IE&IS) department combines disciplinary knowledge from the humanities, social sciences and technical sciences to solve the complex problems of industries and society. We collaboratively focus on and create responsible and effective innovations for the research themes: Humans and Technology, Supply Chain Management, Sustainability and Circularity, and Value of Data-Driven Intelligence.

Information

Do you recognize yourself in this profile and would you like to know more? Please contact the hiring manager Isel Grau ([email protected]), or Yingqian Zhang ([email protected]). 

Visit our website for more information about the application process. You can also contact HR Services ([email protected]).

Curious to hear more about what it’s like as a PhD candidate at TU/e? Please view the video.

Are you inspired and would like to know more about working at TU/e? Please visit our career page.

Application

We invite you to submit a complete application by using the apply button. The application should include a:

• Cover letter in which you describe your motivation and qualifications for the position.
• Curriculum vitae, including a list of your publications and the contact information of three references. Kindly note that we may reach out to references at any stage of the recruitment process. We recommend notifying your references upon submitting your application.

Ensure that you submit all the requested application documents. We give priority to complete applications.

We look forward to receiving your application and will screen it as soon as possible. The vacancy will remain open until the position is filled.

Please note

• You can apply online. We will not process applications sent by email and/or post. 
• A pre-employment screening (e.g. knowledge security check) can be part of the selection procedure. For more information on the knowledge security check, please consult the National Knowledge Security Guidelines.
• Please do not contact us for unsolicited services.