欧美高清

Publications

, ,

EURO Working Group on Locational Analysis (2025)

, ,

EURO 2025: 34th European Conference on Operational Research (2025)

, Fekete Sándor P, Kalcsics J?rg, Kleist Linda

Annals of Operations Research Vol 321, pp. 79-101 (2022)

, Kalcsics J?rg

European Journal of Operational Research Vol 296, pp. 22-43 (2022)

, Fekete Sándor P, Kalcsics J?rg, Kleist Linda

WALCOM 15th International Conference on Algorithms and Computation, WALCOM 2021 Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) Vol 12635 LNCS, pp. 103-115 (2021)

Research Interests

My research interests lie primarily in continuous optimisation, focussing on facility location problems and the use of computational geometry in their solution. Of particular interest are the issues and applications of considerate and competitive location in the presence of barriers to and enablers of movement, investigating game-theoretic interpretations of such problems and utilising machine learning techniques to ascertain optimal locations and strategies.

Professional Activities

Speaker

17/9/2025

Speaker

23/6/2025

Speaker

29/5/2025

Speaker

21/5/2025

Speaker

20/5/2025

Participant

26/4/2025

Projects

Byrne, Thomas (Principal Investigator)

03-Jan-2025 - 30-Jan-2025

Finn, Dominic (Co-investigator) Byrne, Thomas (Principal Investigator)

06-Jan-2023 - 31-Jan-2024

Byrne, Thomas (Principal Investigator) Suzuki, Atsuo (Co-investigator)

The problem of finding optimal locations for service facilities is of strategic importance. The optimal placement of a fire station could save a forest from destruction, that of a hospital could save numerous lives, and that of a supermarket could save thousands of minutes in customers’ travel times. Although an increasing number of models have been proposed, an adequate representation of demand is often crucially neglected. In most models customer demand is assumed to be discrete and aggregated to a relatively small number of points. However, in many applications the number of potential customers can be in the millions and representing every customer residence as a separate demand point is usually infeasible. Therefore it may be more accurate to represent customer demand as continuously distributed over some region.

Furthermore, the region of demand and the region over which a facility can be feasibly located are often assumed to be convex polygons. However, this supposition is not realistic for real-world applications. Moreover, in urban settings the predominantly used straight-line distance does not adequately represent the realised distance for the customer. Further problems arise when we introduce non-traversable areas (e.g. rivers or parks) since this fundamentally alters the way we measure distances between facilities and their demand.

Bringing together results from my PhD and recent work by Professor Suzuki, our goal is to plug the hole in such models by extending existing facility location algorithms to work with this continuous demand distributed over an arbitrary area using a more appropriate measure of distance for real-world applications. This will not only produce more applicable results but also increase the focus of academics onto these very relevant, but unjustifiably ignored, restrictions to such facility location problems, whilst bridging the gaps between academic communities since disciplines such as geometry, calculus, and algebra are also required.

25-Jan-2022 - 24-Jan-2023