Development of Model of Integral Fire Risk Management By Correlation-Registration Analysis

This work is devoted to the construction of a mathematical model for controlling integral fire risk using correlation-regression analysis. On the basis of previous work, work is ongoing on developing a risk management method. For this purpose, the second cluster, which includes the Kharkiv region, was analyzed. In the field of initial data, included statistics for 2010-2017 on parameters such as the number of victims of fire in a year, the time of flight of fire and rescue units to the place of fire, the time of their localization of fire, the time of elimination of the fire and the number of people living in the study area. Using statistical data, the risk of death from a fire in a year was mathematically calculated and entered into the field of output data as a parameter to be investigated.

The study cluster includes 9 areas, therefore 63 events were received and analyzed. As a result, a regression equation was obtained, which makes it possible to predict a risk with a 95% confidence. For example, the Kharkiv region conducted a forecast of the corresponding risk, with the actual result close to the forecasted, and lies within the permissible limits with a deviation of 5%, which indicates the reliability of the developed mathematical model.