Select Page

Manuel Lopez-Ibanez

Senior Lecturer in Decision Sciences and Business Analytics


What are your research interests?

My main research interest is mathematical optimization, in particular mixed-integer programming, metaheuristics, evolutionary algorithms, bayesian optimization, hyper-parameter optimisation (HPO), and the intersection between optimization and machine learning. Within this context, I’m interested in the empirical analysis of optimisation algorithms and their automatic configuration, selection and design. I’m particularly interested in problems with difficult features, such as mixed-integer, multi-objective, dynamic, noisy, computationally expensive, and/or black-box. Such problems arise in a wide range of scenarios, such as logistics, planning, scheduling, timetabling, industrial production, simulation-optimization, robotics, and space exploration.

What is the focus of your current research?

Apart from collaborations with companies, my current research is focused in three main areas:

* The automatic configuration, selection and design of algorithms, in particular, optimisation and machine learning methods. This can be seed as a superset of AutoML and a subset of AutoAI.

* The development and evaluation of optimisation algorithms for problems with multiple conflicting objectives where a human decision-maker is expected to interact with the algorithm. Here, the human factor makes more difficult the benchmarking and analysis of algorithms.

* The resolution of computationally-expensive black-box problems in permutation spaces, where the performance of current Bayesian optimisation approaches is underwhelming.

What are some projects or breakthroughs you wish to highlight?

I am the lead developer of irace (, a tool for the automatic configuration of algorithms.  This tool has been downloaded more than X times and the corresponding paper has more than X citations.

Our recent paper “Reproducibility in Evolutionary Computation”( is the first comprehensive discussions of reproducibility in the context of Evolutionary Computation and metaheuristics optimization. Several of our suggestions in that paper are becoming standard practice in the field.

As far as I am aware, our paper “Incorporating Decision-Maker’s Preferences into the Automatic Configuration of Bi-Objective Optimisation Algorithms”( is the first work that interactively incorporates DM preferences into the automatic configuration of an optimisation algorithm.

Our paper “Machine Decision Makers as a Laboratory for Interactive EMO” ( was among the first to highlight that we need more realistic simulations of decision-makers to evaluate such interactive multi-objective algorithms.  This has motivated a lot of new research in this topic.

What memberships and awards do you hold/have you held in the past?

* Elected member of the Business Committee of GECCO (2024 – 2028) in charge of selecting and advising the General Chair and Editor-in-Chief of the Genetic and Evolutionary Computation Conference (GECCO).

* Chair of the Selection Committee for the ACM SIGEVO Best Dissertation Award (2020 – 2022). The award carries a monetary value of 2,000 USD to be awarded to the winner and 1,000 USD to each of the honourable mentions.

* Member of the IEEE Task Force on Automated Algorithm Design, Configuration and Selection (Oct 2019 – Present)

* Elected member of the ACM SIGEVO Executive Board (Jul 2019 – Present) SIGEVO is the Special Interest Group on Evolutionary Computation of ACM.

I was awarded the “SEIO – BBVA Foundation Award 2021” for the paper “Construct, Merge, Solve & Adapt A new general algorithm for combinatorial optimization” by the Spanish Society of Statistics and Operations Research – BBVA Foundation.

Our team (Martin Zaefferer, Manuel López-Ibáñez and Ekhine Irurozki) won the 1st place in the AI for TSP Competition of the 30th International Joint Conference on Artificial Intelligence (IJCAI-21).

Our paper “Incorporating Decision-Maker’s Preferences into the Automatic Configuration of Bi-Objective Optimisation Algorithms” was selected as one of the Editor’s Choice Articles of the European Journal of Operational Research, January, 2021.

What is the biggest challenge in Data Science and AI right now?

One of the biggest challenges is how to produce solutions that work beyond the assumptions and simplified environments seen in academic papers. Reproducibility, trustworthiness and explainability are also major interconnected challenges that currently hinder adoption.

What real world challenges do you see Data Science and AI meeting in the next 25 years?

Data Science and AI are going to become fundamental tools in all aspects of our society and economy. I expect that AI methods are going to be able to design and self-improve themselves and also come up with innovative designs, mathematical proofs, and scientific theories that are currently beyond our imagination.

Find out more about Manuel’s research at Research Explorer.