The dissertation is devoted to solving an important scientific and practical
task in the field of fire safety - developing a method for the combined deployment
of fire units with different quantitative and qualitative composition in local areas
in order to increase the level of fire safety in such areas with different social and
industrial properties when arranging new fire units and controlling compliance
with the safety requirements of districts, settlements and amalgamated territorial
communities that already have equipped fire units.
The introduction provides a general description of the thesis. The
relevance of the dissertation topic is substantiated, the purpose of the work and
the main tasks of the research are formulated, and the connection of the work with
scientific programmes is shown.
The first section, ANALYSIS OF EXISTING APPROACHES TO THE
DISTRIBUTION OF FIREFIGHTERS WITHIN THE BOUNDARIES OF
AN ADMINISTRATIVE TERRITORIAL UNIT, analyses the regulatory
framework for the deployment of fire departments in cities and amalgamated
territorial communities in Ukraine and establishes that the determining criteria are
population density, travel radius and standard time of arrival at the fire scene. An
analysis of existing methods of fire risk assessment has shown that there are two
approaches to fire risk assessment: quantitative and qualitative. Each approach has
its advantages and disadvantages. However, all modern risk assessment methods
use two main indicators: the probability of a fire and the probability of damage
caused by a fire.In the second section, DEVELOPMENT OF NEURONET MODELS FOR
ASSESSING THE LOCAL TERRITORIAL FIRE HAZARD LEVEL, a
new approach to assessing the fire hazard of local territories of any size, including
individual objects, is proposed. Using neural network technologies, a forecasting
model has been developed that allows estimating the fire hazard of a local area
with a correlation degree of r = 0.767. The resulting neural network is selflearning and can improve the assessment results when new statistical data are
introduced. In the third section, "Development of a Mathematical Model of the
Cumulative Deployment of Fire Units of Different Functional Capacities", it is
shown that the main task of optimising the deployment of fire units in the region
is to create conditions under which the level of fire risk in each point of the region
would correspond to the firefighting capabilities of a potential fire unit, while
minimising financial costs. In the fourth section, METHOD OF AGGREGATE DEPLOYMENT OF
FIRE UNITS OF DIFFERENT FUNCTIONAL CAPACITIES, the proposed
structure of a geographic information system for fire station management consists
of eight layers. The presence of the presented composite layers allows collecting
and analysing information about the nature of the terrain, its buildings, the degree
of fire hazard, the road network and traffic situation, the number and size of fires,
the location of fire and rescue units, the number of fire and rescue forces and
means on the ground. Scientific novelty of the results.
1. For the first time, a methodology has been developed and, on its basis,
neural network models for assessing the level of fire hazard of a local area have
been obtained, which, unlike others, take into account both the parameters of
settlement development in a densely populated area and the spatial distribution of
population and building density, transport and communication network, spatial
distribution of vegetation density and type, and statistical data on landscape fires
for a local area of a large scale, and which allow grading objects and local areas of
the territory by the level of fire hazard.
2. A mathematical model of the aggregate deployment of fire units of
different functional capacities, which takes into account the levels of fire risks of
objects or local territories, has been improved, the objective function of which is
to minimise the number of fire units by reducing the area of intersection of the
service areas of the units and comparing the level of fire hazard of objects with
the functional capacity of all fire units that can be involved in extinguishing a fire
in a normal time with additional restrictions on the condition of a phased increase
in forces t
3. For the first time, a method of aggregate deployment of fire units of
different functional capacities has been developed and implemented in software,
based on the developed mathematical model of aggregate deployment of fire units
of different functional capacities, which, unlike the existing ones, allows
calculating the optimal deployment of future operational service units (fire,
rescue, ambulance, police, citizen assistance centres) in the design of urban
development, development of districts and arrangement of amalgamated territorial
units.
