RUS | UKR | ENG || DonNTU > Master's portal
  Kazakova Julia Kazakova Julia

Faculty: Computer sciences and technologies
Department: Аutomated сontrol systems
Speciality: Information control system and tehnologies

Theme of master's work:
Development of a computer decision support system for fire suppression in inhabited and office buildings
Scientific adviser: Martynenko Tatiana 		 
  About author

 

INTRODUCTION

    Fire Service created to protect the lives and health of citizens, private, collective and state property from fires.
    Management main objectives of fire departments are localization and liquidation of fires, under condition of saving lives and maintenance the minimum economic cost.
    In city conditions firemen meet many difficulties interfering liquidation of burning with suppression of fires. These difficulties include: increased number of tall buildings and high-rise buildings, the increase in industrial areas and civilian facilities, the widespread use of polymeric combustible materials, and the increase in traffic on city streets leading to the increase in the time of following of fire divisions. All these factors contribute to the growth of the number and scales of fires. Consequently, it is necessary to improve the efficiency of management of fire departments.
    One of the most important aspects in work of fire departments is the formation route sheet, which indicates the number of forces and means, depending on the rank of fire, and the optimal route of fire fighting techniques.

URGENCY

    One of the ways to improve the management of fire departments is to improve the quality of information processing and information-analytical work, as necessary preconditions for the development and adoption of scientifically-proved and effective management decisions.
    Application of new information and communication technologies, significantly improves the quality of operational management of fire departments.
    Increasing the number and complexity buildings and increased traffic necessitated the transition from the existing traditional management techniques to the integrated applications of automatic control systems. These automated systems are based on the knowledge of highly qualified professionals and must provide preparation of decisions. In this regard, the development of decision support system (DSS) for management of fire departments is required.
    Automation of the management of forces and means, equipment of fire divisions by computer facilities make it possible to reduce the subjective factor in decision-making.

PURPOSE AND TASKS

    The purpose of master's work is increase of efficiency management decisions in the liquidation of fires in inhabited and office buildings.
    Main tasks:
    - Implement structural and algorithmic analysis of systems management of fire departments;
    - Definition of rank of fire in inhabited and office buildings;
    - Optimization of route selection of fire fighting equipment.
    The object of research – the system support management decisions at the operational management of fire departments.
    Subject of research – methods, tools, models and algorithms that would allow the management of fire departments.
    Research methods – the systems analysis, the management and decision-making theory, methods of expert classification, the theory of fuzzy sets, graph theory, evolutionary computation.

PLANNED SCIENTIFIC NOVELTY

  1. The model representation of highways of city on the basis of graph theory that allowed to determinate the optimum route to fire place is developed.
  2. The method of coding potential solutions in the form of a complex chromosome, that allowed applying the evolutionary approach to reduce the duration of movement of fire equipment and personnel.
  3. A modification of the evolutionary method, for which the problem-oriented operators of selection and crossover are developed, that allowed raising the efficiency of managerial decisions in fire suppression is offered.
  4. The system, which allows determining the necessary structure of forces and means to extinguish the fire and choose the most optimal route of the fire divisions, is developed.

PLANNED PRACTICAL RESULTS

    Planned practical meaningfulness of the work is to develop a computer decision support system for operational management of fire departments for fire suppression in inhabited and office buildings.
    Application of the developed information decision support system will improve the efficiency and quality of management of fire departments, will reduce the influence of subjective factors in making management decisions, improve the management of the fighting. This system provides the operations manager and dispatcher of control center possibility at any time to determine the necessary structure of forces and means to extinguish the fire and choose the best route for techniques and personnel to the fire place.

METHOD USED FOR MAKING DECISIONS AT MANAGEMENT OF FIRE DIVISIONS

    To determine the rank of fire may be used methods of expert estimations, the theory of games, the statistical analysis, neural networks, and theory of fuzzy sets. To determine the rank of fire in inhabited and office buildings are offered to use the apparatus of the theory of fuzzy sets. This approach provides good accuracy for solving the problem, as considered blurred boundaries between adjacent ranks of the fire (the same value may correspond to adjacent ranks of fire with varying degrees of accessory).
    A special feature of the problem finding the optimal path of the fire-fighting equipment is that the best solution is found by the criterion of minimum time.
    To determine the optimal route using methods of full search, branch and bound method, the Dijkstra's algorithm, the algorithm of Floyd–Warshall, evolution computations. Based on the analysis discussed methods of determining the optimal path, we can conclude that for the problem of determining the route of fire equipment and personnel is expedient to apply evolution computations.

DEVELOPMENT OF A COMPUTER DECISION SUPPORT SYSTEM FOR FIRE SUPPRESSION IN INHABITED AND OFFICE BUILDINGS

    An important step in responding to the message of fire is the correct definition of rank of fire, which allows installing the necessary forces and means to liquidate burning.
    Rank of fire is conditional sign of complexity of the fire, which determines in the departure schedule necessary forces and means involved in fire fighting. Number of fires ranks the techniques sent on each rank is established by the chief of garrison of fire service in the departures schedule of typical techniques according to available resources and characteristics of objects located on the territory of the garrison.
    Many factors, characterizing the situation on the burning object, are considered at definition of rank of fire, i.e. the rank of fire depends on many parameters Q:

R = F(Q1,Q2,…,Qm), (1)

    where Qi – parameters, that influence on rank of fire;
    m – number of parameters, that influence on rank of fire.
    Then the number and structure of forces and means used at fire suppression is a function of the rank:

S=F(R), (2)

    where S – the vector with the following structure:

S=(St1, K1, St2, K2,…, Stn, Kn), (3)

    where Sti – kind of fire fighting equipment, means of suppression;
    Ki – quantity of fire fighting equipment;
    n – quantity of kinds of fire fighting equipment and technical means.
    To determine the rank of fire in inhabited and office buildings are offered to use the apparatus of the theory of fuzzy sets. This approach provides good accuracy for solving the problem, as considered blurred boundaries between adjacent ranks of the fire (the same value may correspond to adjacent ranks of fire with varying degrees of accessory).

    It is necessary to determine the optimal route to fire place, which provides the minimum arrival time, as at liquidation of fires time is the determinative influencing scales of consequences.
    The route is sequence of streets and alley between a fire station and the fire place. Therefore, it is offered to present route as set of crossroads which the fire-fighting divisions should drive, i.e. In the form of vector X = (p0, p1, p2, ..., pn), where
    p0 – the starting crossroads (the location of the fire station);
    pn – the final crossroads (fire place);
    pi – road crossroads, (i ∈ 1, n);
    n – quantity of crossroads on way.
    Rationale for the route is carried out based on the criterion of minimization the arrival time of personnel and fire-technical equipment to the place of fire. At travel time influence many factors: the length of the path, quantity of crossroads, traffic.
    Consequently, we must find route X on which movement time will be minimal.

    The model representation of the city highways proposed to be constructed on the basis of graph theory. Information on the highways of the city is represented as follows: crossroads of streets is vertices, and streets and alleys is edges.
    To determine the optimal route of fire equipment and personnel are offered to use evolutionary computation. They provide good accuracy and execution time.
    For application of evolutionary algorithm to this problem it is necessary to define:
    - Structure of the chromosome;
    - Generate the sample;
    - The selection operator;
    - The crossover operator;
    - The mutation operator;
    - Strategy selection.
    The chromosome (individual) is route of fire-fighting equipment from fire station to fire place.
    Fitness function is movement time on route.
    The selection operator is based on the "roulette wheel". Individual with a smaller value of fitness-function has higher probability to be selected.
    For a pair of selected chromosomes with a probability of Vc (Vc for a given problem is selected in the range [0,5 .. 0,8]) crossover operator (creation offspring) is carried out. Figure 1 shows a graphical representation of the crossover operator.
Figure 1. Presentation crossover operator
Figure 1. Presentation crossover operator
(Animation: volume – 52,4 KB; size – 645х337; amount of shots – 4; amount of repeated cycles – a continuous cycle of reiteration)

    For the offspring with probability Vm (Vm for a given problem is selected in the range [0,001 .. 0,01]) must execute the mutation operator that helps to avoid falling into local optimum.
    Population is expanding by adding new individuals. To reduce the population to its original size is proposed to use elite selection. Elite individuals represent 10% of the total sample.
    The algorithm holds when one of the stopping criteria is met:
    1) population consists of identical elements;
    2) achieve a given accuracy, for all chromosomes difference in values of fitness-function does not exceed value ε;
    3) achievement of a certain age population (number of epochs).
    In general, the structure of a compute decision support system for fire suppression in inhabited and office buildings may be represented as a set of modules (Figure 2).

Figure 2. The structure of decision support in fire suppression in inhabited and office buildings
Figure 2. The structure of decision support in fire suppression in inhabited and office buildings

CONCLUSIONS

    Application of automated systems for management divisions of fire protection, forces and means allow to reduce influence of the subjective factor at development decisions, raise management efficiency operations, help to develop scientifically decisions and to raise quality of management of forces and means on fire. Fast and correct decision-making will significantly reduce the economic costs of fire suppression and quantity of victims.
    In the master's work to the actual task of management of fire departments in fire suppression is considered. Scientific research is executed to get an idea about the current state of problem, the existing software tools for management of fire departments are analyzed, various methods for determining the rank of fire in inhabited and office buildings, and also methods for determining the optimal route are considered.
    For definition of rank of fire it is the most expedient to use the apparatus of fuzzy sets. For definition of optimum route of fire techniques and personnel it is offered to use evolutionary calculations.

REFERENCES

  1. Тетерин И.М., Климовцов В.М., Прус Ю.В. Методология разработки экспертных систем для оперативного управления пожарными подразделениями / Научный интернет портал «Технологии и системы безопасности» // Интернет-журнал: Технологии техносферной безопасности – 2008. – № 5. [Электронный ресурс]: Режим доступа: URL: http://ipb.mos.ru/ttb
  2. Тетерин И.М., Топольский Н.Г., Климовцов В.М., Прус Ю.В. Применение систем поддержки принятия решений руководителями оперативных подразделений при тушении пожаров в крупных городах / Научный интернет портал «Технологии и системы безопасности» // Интернет-журнал: Технологии техносферной безопасности – 2008. – № 4. [Электронный ресурс]: Режим доступа: URL: http://ipb.mos.ru/ttb
  3. Про затвердження Тимчасового порядку організації внутрішньої, гарнізонної та караульної служб МНС України. Наказ МНС України № 794 від 31.10.2008 р.
  4. Палюх В. Г. Облік і аналіз пожеж в Україні. Нормативні акти / Харків – 2004. –83 с.
  5. Тетерин И.М. Теоретико-игровые методы в системах поддержки принятия решений для руководителя тушения пожара / Научный интернет портал «Технологии и системы безопасности» // Интернет-журнал: Технологии техносферной безопасности – 2008. – № 5. [Электронный ресурс]: Режим доступа: URL: http://ipb.mos.ru/ttb
  6. Моргун О.М., Моргун Л.О. Комп’ютерна система оптимізації вибору маршрутів слідування аварійно-рятувальної техніки / Национальная библиотека имени В. И. Вернадского // Пожежна безпека: теорія і практика. Збірник наукових праць. – Черкаси: АПБ, 2008. – № 1. [Электронный ресурс]: Режим доступа: URL: http://www.nbuv.gov.ua/portal/natural/Pbtp/texts/2008-1/st_17%20morgun.pdf
  7. Харитонова, Е. В. Графы и сети: учебное пособие / Е. В. Харитонова. – Ульяновск: УлГТУ, 2006. – 92 с.
  8. Скобцов Ю. А. Основы эволюционных вычислений: Учебное пособие. / – Донецк: ДонНТУ, 2008. – 326 с.
  9. Снитюк В.Е., Джулай А.Н. Эволюционный метод определения кратчайшего пути проезда пожарного расчета к месту пожара с оптимизированным пространством поиска. XII-th International Conference Knowledge-Dialogue-Solution June 20-25, 2006, Varna (Bulgaria). / ARNIT – Информационные Интеллектуальные Системы. [Электронный ресурс]: Режим доступа: URL: http://artint.com.ua/pdf/Russian/evolfireway_rus.pdf
  10. Требнев В. В. Справочник руководителя пожара. Тактические возможности пожарных подразделений / М., «ПожКнига», 2004. – 256 с.
  11. Рутиковская Д., Пилиньский М., Рутиковский Л. Нейронные сети, генетические алгоритмы и нечеткие системы: Пер. с польск. И. Д. Рудинскго / М.: Горячая линия – Телеком, 2006. – 452 с.

While writing the given abstract the master's work has not been completed yet. The final date of the work completed is December, 2010. The text of master's work and materials on this topic can be received from the author or her research guide after the indicated date.

© DonNTU 2010, Kazakova Julia
  DonNTU > Master's portal > About author