Semnan University Press Journal of Modeling in Engineering 2008-4854 Articles in Press 2026 01 18 A hybrid model for evaluating the performance of computer network sensors using Data Envelopment Analysis and Metaheuristic Algorithms A hybrid model for evaluating the performance of computer network sensors using Data Envelopment Analysis and Metaheuristic Algorithms 10431 10.22075/jme.2026.38435.2877 FA Mohammad Sadegh Behruz PhD in Industrial Management, Faculty of Management and Economics, Tehran University of Science and Research, Tehran, Iran Meysam Karamipour Postdoctoral, Iran University of Science and Technology, School of Management, Economics and Progress Engineering, Iran Rahman Noormohammadi Ph.D. student of Industrial Management, Faculty of Management and Economics, Branch of Science and Research, Islamic Azad University, Tehran, Iran Journal Article 2025 07 24 Storing and processing information obtained from sensors and sensors, as well as using wireless and real-time prediction systems and data analysis, creates added value. Therefore, evaluating the performance of sensors in computer networks is important. The purpose of this research is to present a model for evaluating the performance of sensors using mathematical models and optimizing the results using simulation tools. The research data is related to a computer network simulated in the OPNET environment. This computer network is designed based on sensors and sensors that are placed in the gas pipeline in order to identify and diagnose critical defects and maintenance program data. In the present research, the problem is modeled using the mathematical model BCC and SBM of data envelopment analysis hyper-efficiency, and after identifying efficient units, the performance efficiency values are optimized with the MLP neural network and its combination with the cuckoo algorithm. To measure the accuracy of the model performance, the parameters of mean square error, correlation coefficient, standard deviation and mean absolute error have been evaluated. The results of the research indicate that these values are 0.2605, 0.123666, 0.66 and 0.89853 in the neural network, and 0.1037, 0.03462222, 0.43 and 0.94829 in the hybrid algorithm, respectively. Accordingly, the use of the hybrid algorithm improves the data learning process in the network and increases the accuracy in the final outputs of the research model, Storing and processing information obtained from sensors and sensors, as well as using wireless and real-time prediction systems and data analysis, creates added value. Therefore, evaluating the performance of sensors in computer networks is important. The purpose of this research is to present a model for evaluating the performance of sensors using mathematical models and optimizing the results using simulation tools. The research data is related to a computer network simulated in the OPNET environment. This computer network is designed based on sensors and sensors that are placed in the gas pipeline in order to identify and diagnose critical defects and maintenance program data. In the present research, the problem is modeled using the mathematical model BCC and SBM of data envelopment analysis hyper-efficiency, and after identifying efficient units, the performance efficiency values are optimized with the MLP neural network and its combination with the cuckoo algorithm. To measure the accuracy of the model performance, the parameters of mean square error, correlation coefficient, standard deviation and mean absolute error have been evaluated. The results of the research indicate that these values are 0.2605, 0.123666, 0.66 and 0.89853 in the neural network, and 0.1037, 0.03462222, 0.43 and 0.94829 in the hybrid algorithm, respectively. Accordingly, the use of the hybrid algorithm improves the data learning process in the network and increases the accuracy in the final outputs of the research model,