Designing an Integrated Framework Based on Explainable Artificial Intelligence and Data Envelopment Analysis for Evaluating the Performance of Operational Units of the South Zagros Oil and Gas Production Company

Seyed Morteza  Mousavi; Javad Gerami; Mohammadreza  Mozaffari; Roya  M. Pour Ahari; Mohammadreza  Feylizadeh

Authors

Seyed Morteza Mousavi Department of Industrial Engineering, Na.C., Islamic Azad University, Najafabad, Iran Author https://orcid.org/0009-0000-1526-5798
Javad Gerami Department of Industrial Engineering, Shi.C., Islamic Azad University, Shiraz, Iran https://orcid.org/0000-0001-6829-1412
Mohammadreza Mozaffari Department of Industrial Engineering, Shi.C., Islamic Azad University, Shiraz, Iran Author https://orcid.org/0000-0003-3160-271X
Roya M. Pour Ahari Department of Industrial Engineering, Na.C., Islamic Azad University, Najafabad, Iran Author https://orcid.org/0000-0002-8952-8682
Mohammadreza Feylizadeh Department of Industrial Engineering, Shi.C., Islamic Azad University, Shiraz, Iran Author https://orcid.org/0000-0002-1382-7328

Keywords:

Data Envelopment Analysis, Explainable Artificial Intelligence, Machine Learning, SHAP, Operational Efficiency, Oil and Gas Industry, XGBoost

Abstract

This study aimed to design and evaluate an integrated framework combining Data Envelopment Analysis (DEA), machine learning, and Explainable Artificial Intelligence (XAI) for assessing, predicting, and interpreting the operational efficiency of oil and gas production units. The study employed an applied quantitative design using 30 decision-making units (DMUs) evaluated for the 2026–2027 operational year. Four inputs comprising feedstock, energy consumption, water consumption, and labor; two desirable outputs comprising main production and revenue; and five undesirable outputs comprising CO₂, NOx, SO₂, PM10, and COD were analyzed. DEA was first used to calculate relative efficiency scores and identify efficient and inefficient units. These efficiency scores were then used as the target variable in supervised machine-learning models, including linear regression, support vector regression, random forest, gradient boosting, and XGBoost. Model performance was evaluated using coefficient of determination, mean absolute error, root mean square error, and cross-validation. SHapley Additive exPlanations (SHAP) were subsequently applied to interpret the selected predictive model and determine the contribution of each operational variable to efficiency. The mean DEA efficiency score was 0.879, indicating an average efficiency gap of 12.1% relative to the estimated frontier. Five DMUs achieved an efficiency score of 1.000 and were classified as efficient. XGBoost demonstrated the best predictive performance, with (R²=0.944), MAE=0.017, and RMSE=0.026; its cross-validated (R²) was 0.889 with RMSE=0.034. SHAP analysis identified energy consumption as the most influential predictor of efficiency (18.6%), followed by main production (16.4%), CO₂ emissions (14.1%), revenue (12.3%), and labor (10.5%). Higher production and revenue generally increased predicted efficiency, whereas excessive energy use, pollutant emissions, and disproportionate labor utilization reduced it. The integrated DEA–machine learning–XAI framework provided a multidimensional, predictive, and interpretable approach to operational performance evaluation.

References

Agrawal, R. (2026). BQEB ForecastBench: Benchmarking AI Models for Smart Grid Forecasting Using BQEB-Data V1. https://doi.org/10.21203/rs.3.rs-10484554/v1

Ahmed, S., & Shahzad, K. (2022). Augmenting Business Process Model Elements With End-User Feedback. IEEE Access, 10, 115635-115651. https://doi.org/10.1109/access.2022.3216418

Ajayi, O. O., Kurien, A., Djouani, K., & Dieng, L. (2025). A Proactive Predictive Model for Machine Failure Forecasting. Machines, 13(8), 663. https://doi.org/10.3390/machines13080663

Akram, K., Bhutta, M. U., Butt, S. I., Rizwan, M., Khan, D. M. B., Khan, M., & Khan, A. (2026). A Novel Multi-Objective Dynamic Flexible Job Shop Scheduling Algorithm Using Reinforced Learning Based Black Widow Spider Algorithm. PLoS One, 21(4), e0347108. https://doi.org/10.1371/journal.pone.0347108

Ali, A., Qamar, R., Asif, R., & Hina, S. (2026). Benchmarking Energy Efficiency of Supervised Machine Learning Models on Multi-Domain Classification Datasets. Information, 17(7), 652. https://doi.org/10.3390/info17070652

Álvarez‐Díez, S., Baixauli‐Soler, J. S., & Kondratenko, A. (2025). Evaluating Expert Decision Systems for Exchange Rate Insurance. International Journal of Business Analytics, 12(1), 1-25. https://doi.org/10.4018/ijban.378390

Baqer, M. (2026). A Machine Learning-Centric Taxonomy and Structured Characterization of Public Datasets for Upstream Oil and Gas. Big Data and Cognitive Computing, 10(6), 188. https://doi.org/10.3390/bdcc10060188

Champa, S. S., & Segall, R. S. (2026). Artificial Intelligence and Machine Learning Framework for Smart Manufacturing Optimization. International Journal of Artificial Intelligence, 2(1), 1-46. https://doi.org/10.4018/ijaibm.410302

Crăciun, R.-A., Pietraru, R. N., & Moisescu, M. A. (2024). Internet of Things Platform Benchmark: An Artificial Intelligence Assessment. Revue Roumaine Des Sciences Techniques — Série Électrotechnique Et Énergétique, 69(1), 97-102. https://doi.org/10.59277/rrst-ee.2024.1.17

Elrefaie, M. (2025). CarBench: A Comprehensive Benchmark for Neural Surrogates on High-Fidelity 3D Car Aerodynamics. https://doi.org/10.48550/arxiv.2512.07847

Forniés-Tabuenca, D., Uribe, A., Otamendi, U., Artetxe, A., Rivera, J. C., & Lacalle, O. L. d. (2026). REMoH: A Reflective Evolution of Multi-Objective Heuristics Approach via Large Language Models. https://doi.org/10.21203/rs.3.rs-8837965/v1

Ibrahim, M. E. A., Ahmed, A. E. S., & Daadaa, Y. (2026). Beyond Neural Solvers: A Critical Review of Machine Learning for Combinatorial Optimization. Mathematics, 14(12), 2208. https://doi.org/10.3390/math14122208

Jung, H., & Yoo, P. D. (2025). SMART: Structured Missingness Analysis and Reconstruction Technique for Credit Scoring. Scientific reports, 15(1). https://doi.org/10.1038/s41598-025-99997-4

Lee, D., & Choi, B.-S. (2026). A Structure‐Based Benchmarking Suite for Quantum Learning. Concurrency and Computation Practice and Experience, 38(10). https://doi.org/10.1002/cpe.70758

Mannan, K. A. (2026). Comparative Evaluation of Traditional Machine Learning and Transformer Models for Fake News Detection. https://doi.org/10.21203/rs.3.rs-10953606/v1

Nimmala, R. (2023). Enhancing Financial Risk Management: Utilizing Machine Learning in Climate Risk Model Benchmarking. Journal of Mathematical & Computer Applications, 2(1), 1. https://doi.org/10.47363/jmca/2023(2)146

Orazov, B., Nobatov, A., & Mamiyev, A. (2026). Application of Machine Learning Models for Predictive Maintenance in Industrial Production Systems. 10. https://doi.org/10.1117/12.3109286

Patel, D. (2026). Adversarial Intelligence: A Systematic Review of AI-Driven Cyber Threat Detection, Adaptive Malware Evasion, and Arms Race Dynamics in Zero Trust Security Environments. https://doi.org/10.21203/rs.3.rs-10137205/v1

Rahaman, R. (2025). Automated Deployment and Performance Benchmarking of Machine Learning Workloads on Hadoop Clusters Using Ansible. https://doi.org/10.21203/rs.3.rs-7003490/v1

Raynal, J., Slangen, P., Raynal, E., & Margerit, J. (2026). Detecting Explanatory Insufficiency in Learned Representations: A Framework for Representational Vigilance. https://doi.org/10.48550/arxiv.2606.13172

Rharif, A., Charafi, Z., Zaydi, M., Belkhala, S., & Maleh, Y. (2026). Deployable AI for IoT/IIoT Security: A Systematic Review of Labeling, Transfer, Resource, and Explainability Constraints. https://doi.org/10.21203/rs.3.rs-9929399/v1

Rokade, M. D. (2024). Advancements in Privacy-Preserving Techniques for Federated Learning: A Machine Learning Perspective. Journal of Electrical Systems, 20(2s), 1075-1088. https://doi.org/10.52783/jes.1754

Scheinert, D., Becker, S., Bader, J., Thamsen, L., Will, J., & Kao, O. (2022). Perona: Robust Infrastructure Fingerprinting for Resource-Efficient Big Data Analytics. 209-216. https://doi.org/10.1109/bigdata55660.2022.10020860

Singhal, S. (2026). AI-driven Cybersecurity for Industrial Internet of Things: Architectures, Challenges, Datasets, and Future Research Directions. Frontiers in Big Data, 9. https://doi.org/10.3389/fdata.2026.1938279

Sucuoğlu, B. Y., Beyca, Ö. F., & Kosanoglu, F. (2026). Forecasting Intermittent Sales in Fashion Retail: A Two-Stage Machine Learning Approach. Forecasting, 8(4), 56. https://doi.org/10.3390/forecast8040056

Wan, J., Yar, K. P., Low, M. Y. H., Xu, C., Doan, N. C. N., Ng, H. Y., & Wang, W. (2026). PI-FSL: Physics-Informed Few-Shot Domain Adaptation for Robust Cross-Domain Condition Monitoring. Technologies, 14(3), 167. https://doi.org/10.3390/technologies14030167

Wang, P., & Yu, Z. (2023a). RayBench: An Advanced NVIDIA-Centric GPU Rendering Benchmark Suite for Optimal Performance Analysis. Electronics, 12(19), 4124. https://doi.org/10.3390/electronics12194124

Wang, P., & Yu, Z. (2023b). RenderBench: The CPU Rendering Benchmark Suite Based on Microarchitecture-Independent Characteristics. Electronics, 12(19), 4153. https://doi.org/10.3390/electronics12194153

Yang, A. I., Woo, J., Zhang, R., Mach, A., Ramkumar, P. N., & Ma, Y. (2025). Tool-Wielding Language-Model-Based Agent Offers Conversational Exploration of Clinical Tabular Data. https://doi.org/10.64898/2025.12.01.25341392

Zakaria, A. (2026). A Human-Centric Fuzzy Decision Support System for Medical Diagnosis Using Fuzzy Cognitive Maps. Scientific reports, 16(1). https://doi.org/10.1038/s41598-026-51590-z

Zewail, R., & Mokhtar, B. (2026). Resource‐Aware Contrastive Scattering Meta‐Learning for Efficient Few‐Shot Acoustic Anomaly Detection. Advanced Intelligent Systems, 8(9). https://doi.org/10.1002/aisy.202501454

Zou, Q. (2025). FML-bench: Benchmarking Machine Learning Agents for Scientific Research. https://doi.org/10.48550/arxiv.2510.10472