Year: 2024
Type: article
Abstract: The article presents a new approach to risk management of emergency situations at cellular network base stations. The diversity of emergencies arising on heterogeneous equipment under the influence of various external factors makes risk management a critically important task. The novelty lies in the creation of a mathematical model that accounts for this diversity, providing more accurate and comprehensive prediction of emergency situations. The model is based on a Bayesian network and generates real-time solutions in the form of probabilities of emergency occurrences, identifying critical points and potential threats to the base station's overall functionality. This helps generate recommendations for reducing risks, identifying priority areas for implementing improvements, and modernizing equipment. To ensure effective interaction with the model, an API is being developed and studied using FastAPI and Python. The API interacts with the Bayesian model created in BayesFusion GeNIe. The model implements a new method for integrating the developed Bayesian network with existing applications based on REST API principles, thus introducing a new approach to risk management. The article describes the processes of API creation, performance testing, and deployment. As a result of using the API, real-time risk management becomes possible, helping operators prevent emergency situations. The integrated model was developed for research purposes to monitor the risk landscape of cellular network base stations. The application of this model significantly increases the level of automation in the risk management process during the operation of cellular network base stations.
Authors: K. E. Grigoriev, V. S. Kanev, A. N. Poletaikin
Language: English
Topic: Risk and Safety Analysis
Subfield: Statistics, Probability and Uncertainty
Field: Decision Sciences
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W4405498033