Fairness in maximal covering location problems

This paper provides a mathematical optimization framework to incorporate fairness measures from the facilities’ perspective to discrete and continuous maximal covering location problems. The main ingredients to construct a function measuring fairness in this problem are the use of (1) ordered weight...

Full description

Saved in:
Bibliographic Details
Published in:Computers & operations research Vol. 157; p. 106287
Main Authors: Blanco, Víctor, Gázquez, Ricardo
Format: Journal Article
Language:English
Published: Elsevier Ltd 01.09.2023
Subjects:
Facility location
Fair resource allocation
Mixed integer non linear programming
Ordered weighted averaging problem
Facility location
90B85
Mixed integer non linear programming
90C11
90B80
90C59
Ordered weighted averaging problem
Fair resource allocation
ISSN:0305-0548
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper provides a mathematical optimization framework to incorporate fairness measures from the facilities’ perspective to discrete and continuous maximal covering location problems. The main ingredients to construct a function measuring fairness in this problem are the use of (1) ordered weighted averaging operators, a popular family of aggregation criteria for solving multiobjective combinatorial optimization problems; and (2) α-fairness operators which allow generalizing most of the equity measures. A general mathematical optimization model is derived which captures the notion of fairness in maximal covering location problems. The models are first formulated as mixed integer non-linear optimization problems for both the discrete and the continuous location spaces. Suitable mixed integer second order cone optimization reformulations are derived using geometric properties of the problem. Finally, the paper concludes with the results obtained from an extensive battery of computational experiments on real datasets. The obtained results support the convenience of the proposed approach. •We define a new fairness measure combining OWA and α-fairness operators.•We prove that our operator fits into the axiomatic of fair operators.•We develop a general mathematical programming model to incorporate that measure into different MCLPs.•We derive MISOCO reformulations for the problem.•We validate our proposal on an extensive battery of computational experiments.
ISSN:0305-0548
DOI:10.1016/j.cor.2023.106287