Resilience is critical in complex industrial systems such as oil, gas, and petrochemical companies, where disruptions can significantly affect performance. This study introduces the Resilience-based Performance Evaluation Model (RPEM), a novel framework for assessing system performance by integrating three Key Performance Indicators (KPIs): Reliability, efficiency, and Specific Energy Consumption (SEC). Unlike previous studies that evaluate these indicators separately, the RPEM holistically combines them to calculate the Resilience Threshold (RT), enabling a comprehensive assessment of system resilience. The proposed model is demonstrated through a case study in a gas refining company. The calculated RT values at the unit level indicate that units C2 and Utility are categorized as ineligible, meaning they are inefficient and non-resilient. It was found that the most important factor causing the undesired performance of the C2 Recovery unit is the failure of the cold box equipment, which has resulted in a decrease in the unit's RT indicator. Response Surface Methodology (RSM) optimizes and forecasts key indicators. The predicted R² of 0.9491 and the Adjusted R² of 0.9874 indicate a strong model fit, with significant regression terms for the RTs, SEC, Efficiency, and Reliability (P-value < 0.0001), demonstrating that high reliability and efficiency, along with near-baseline energy efficiency, optimize system flexibility. This research addresses the lack of a comprehensive resilience evaluation framework, offering actionable insights to improve system performance and flexibility
