34th European Conference on Operational Research - EURO 2025, Leeds (United Kingdom). 22-25 June 2025
Summary:
Although interpretability is gaining traction in Artificial Intelligence and Machine Learning, it remains largely overlooked in Optimization, where black-box models are the standard. We introduce a method to enhance the interpretability of optimization models by focusing on the most influential variables and constraints, effectively reducing complexity with controlled impact on accuracy.
Our approach systematically eliminates extraneous variables and redundant constraints using filters inspired by existing presolve techniques. To ensure feasibility, constraints are carefully adjusted while redundant elements are removed.
We validated this method on 60 linear problems from the GAMS library and openTEPES, a real-world optimization model for long-term energy system expansion. Sensitivity analysis across different thresholds demonstrated that our approach significantly reduces problem size while preserving optimality and feasibility. While preserving the essential structure of the problem, the analysis of the automatic simplication process makes optimization models more transparent and easier to interpret.
Spanish layman's summary:
Presentamos un método para mejorar la interpretabilidad de los modelos de optimización. En lugar de intentar explicar el modelo completo, buscamos una simplifación que contenga las variables y restricciones más importantes, aunque sacrifiquemos un poco de precisión del modelo.
English layman's summary:
We introduce a method to enhance the interpretability of optimization models Instead of trying to explain the entire model, we zero in on the most important variables and constraints — the ones that really drive the results. This approach simplifies the model without sacrificing much accuracy.
Keywords: Programming, Linear, Analytics and Data Science
Publication date: 2025-06-22.
Citation:
B. Ruiz-Gonzalez, S. Lumbreras, J. García-González, Enhancing optimization interpretability through problem simplification, 34th European Conference on Operational Research - EURO 2025, Leeds (United Kingdom). 22-25 June 2025.