Publisher
University of Tennessee at Chattanooga
Place of Publication
Chattanooga (Tenn.)
Abstract
Uncertainty in input data significantly affects the quality of solutions in mathematical optimization, making it crucial to explore alternative decisions when the selected decision becomes unavailable or suboptimal. This study introduces a novel risk-averse decision-making approach for cases where objective function coefficients are uncertain in multi-objective combinatorial optimization problems. We construct a region in the objective space based on reference solutions obtained from the deterministic formulation. Alternative decisions, identified using a neighboring structure that falls within this region, are used to determine risk-preference solutions. We propose two sets of indices to quantify the quality of outcomes and neighboring decisions in terms of performance and risk level. The approach is demonstrated in diverse test cases, highlighting its effectiveness in improving risk-averse decision-making under uncertainty.
Keywords: Multi-objective Combinatorial Optimization, Neighboring Decisions, Risk-averse Decision-Making, Sensitivity Region, Uncertainty.
Document Type
posters
Language
English
Rights
http://rightsstatements.org/vocab/InC/1.0/
License
http://creativecommons.org/licenses/by/4.0/
Recommended Citation
Aththanayake, Chathuri M. and Weerasena, Lakmali, "Risk-Averse Decision-Making in Multiobjective Combinatorial Optimization". ReSEARCH Dialogues Conference proceedings. https://scholar.utc.edu/research-dialogues/2025/posters/2.
Risk-Averse Decision-Making in Multiobjective Combinatorial Optimization
Uncertainty in input data significantly affects the quality of solutions in mathematical optimization, making it crucial to explore alternative decisions when the selected decision becomes unavailable or suboptimal. This study introduces a novel risk-averse decision-making approach for cases where objective function coefficients are uncertain in multi-objective combinatorial optimization problems. We construct a region in the objective space based on reference solutions obtained from the deterministic formulation. Alternative decisions, identified using a neighboring structure that falls within this region, are used to determine risk-preference solutions. We propose two sets of indices to quantify the quality of outcomes and neighboring decisions in terms of performance and risk level. The approach is demonstrated in diverse test cases, highlighting its effectiveness in improving risk-averse decision-making under uncertainty.
Keywords: Multi-objective Combinatorial Optimization, Neighboring Decisions, Risk-averse Decision-Making, Sensitivity Region, Uncertainty.