This article presents the development of a mathematical model for calculating the reliability parameters of smart home system sensors. The primary objective of the study is to develop and analyze a model that can predict the failure probability and uptime of sensors based on statistical and experimental data. To solve this problem, mathematical modeling methods, probability theory, and modern artificial intelligence approaches genetic algorithms (GA) and recurrent neural networks (RNN, LSTM) are applied. These methods improve the accuracy of failure prediction and allow for the consideration of external factors such as temperature, humidity, power consumption, and network load. As a result, a comprehensive model was constructed that describes the behavior of sensors under dynamic operating conditions. A comparative analysis of prediction accuracy was conducted, and key reliability metrics failure rate, mean time between failures, and system availability were evaluated. The developed model can be used in the design, optimization, and maintenance of smart home systems, as well as in other areas of the Internet of Things (IoT) where high reliability of sensor nodes is important.

Reliability, mathematical model, sensors, smart home, genetic algorithm

Singh, K., et al. Reliability on the Internet of Things with designing. Frontiers in Computer Science, 2024.

Zhao, G., et al. Reliability analysis of IoT systems with competitions from Cascading Probabilistic Function Dependence. 2020.

Pang, B., and Evgeny S. Abramov. Reliability Analysis and Parameter Selection for IoT Communication Based on Deep Learning. 2025.

Ergun, K., et al. Simulating Reliability of IoT Networks with RelIoT. 50th Annual IEEE/IFIP International Conference on Dependable Systems and Networks - Supplemental Volume (DSN-S). 2020.

Singh, K., et al. Techniques in reliability of internet of things (IoT) (2025).

Pal, V., Singh, G., Yadav, R. P. Genetic Algorithm Based Method for Analyzing Reliability State of Wireless Sensor Network. Advances in Intelligent Systems and Computing, vol 199. Springer, Berlin, Heidelberg. 2013. doi.org/10.1007/978-3-642-35314-7_66.

Attea, B. A., Hameed, S. M. A Genetic Algorithm for Minimum Set Covering Problem in Reliable and Efficient Wireless Sensor Networks. Iraqi Journal of Science, 2014, Vol 55, No.1, pp:224-240.

Khujamatov, H., et al. ERIRMS: Evaluation of the Reliability of IoT-Aided Remote Monitoring Systems of Low-Voltage Overhead Transmission Lines. Sensors 2024, 24(18), 5970, doi.org/10.3390/s24185970.

Linard, A., Bucur, D., Stoelinga, M. Fault Trees from Data: Efficient Learning with an Evolutionary Algorithm. 2019. DOI:10.48550/arXiv.1909.06258

A.A.Muradova, “Modeling of decision-making processes to ensure sustainable operation of multiservice communication network”, ITB Journal, ISSN: 2337-5787, Vol. 13, No.1, pp.50-62, 2019.

Lam, H. T., Szeto, K. Y. Optimization of Reliability of Network of Given Connectivity using Genetic Algorithm. Physics and Society; Neural and Evolutionary Computing (cs.NE); Social and Information Networks. 2014.

Heidari, E., and Movaghar, A. An Efficient Method Based on Genetic Algorithms to Solve Sensor Network Optimization Problem. International Journal on Applications of Graph Theory in Wireless Ad Hoc Networks and Sensor Networks 3(1), 2011.

A.A. Muradova, and D.T. Normatova, “The results of a research of the reliability of the multiservice communication network”, ICISCT 2022 Applications, trends and opportunities, TUIT, 28-30 September, 2022.

Sofge, D. A., and Elliott, D. L. Improved Neural Modeling of Real-World Systems Using Genetic Algorithm Based Variable Selection. Neural and Evolutionary Computing. 2007.

Subair P. H., Basheer P. I., and Shajil Ameer V. V. “Reliability Modeling for Sensor Systems”, IARJ (International Journal of Research and Analytical Reviews), Vol. 1, Issue 2, 2021.

A.A. Muradova, and D.T. Normatova, “Results of simulation modeling of technical parameters of a multiservice network”, Telkomnika (Telecommunication Computing Electronics and Control),21(3), pp. 702–710, 2023.