Generalized robust goal programming model

Abstract

This study proposes a concise and generalized robust goal programming (RGP) model that simultaneously considers three types of goal functions – right-side penalties, left-side penalties, and both-side penalties – under uncertainties on both the left-hand side and right-hand side. It integrates common uncertainty sets for a comprehensive goal programming model. Experimental results reveal that our model consistently outperforms existing RGP models by incurring fewer penalties, demonstrating enhanced resilience and robustness. This advantage becomes evident when problem coefficients such as costs, profits, and human resource requirements deviate significantly from their default target levels due to real-world conditions. The proposed model not only extends the robustness of traditional goal programming and weighted fuzzy goal programming but also offers improved risk management across various practical scenarios.