An artificial intelligence model to support management decisions to increase the efficiency of preventive measures to prevent fire
Abstract
An artificial intelligence model to support management decisions to increase the efficiency of preventive measures to prevent fire
Incoming article date: 06.03.2024The proposed work considers two types of neural network models for describing fire risks depending on the size of the population and preventive measures. Neural network models make it possible to consider fire risks integrally, taking into account the type of municipality or separately for each of the three existing types. Based on these models, a response surface for fire risks - population and prevention has been implemented, which allows you to assess the magnitude of risks based on input data to optimize decisions made. For a given value of the standard risk, the dependence of optimal prevention was obtained depending on the number of deposits in the municipality, which makes it possible to guarantee fire risks less than or equal to the standard indicators. The article analyzes and evaluates the effectiveness of preventive measures using neural networks. The input data for training the neural network includes fire data collected in Vietnam and Russia (population, number of fires, number of deaths, number of preventive measures). Based on these indicators, the effectiveness of preventive measures is predicted. Based on the forecasting results, decisions can be made to ensure fire safety in the state. The results obtained indicate the possibility of predicting the absolute value of the effectiveness of preventive work based on quantitative and categorical variables. A relatively large forecast error is associated, on the one hand, with the need to take into account a larger number of input parameters, and on the other hand, with the need to increase the size of the neural network training base. After refining the model, the results obtained allow us to evaluate the effectiveness of preventive measures for provinces and cities.
Keywords: fire safety, preventive measures, neural network, management decision support, prediction model