The methods of analytical data processing used for support information tasks for managing the security of territories are considered. The problem of data availability for operating analytics is investigated. The use of end-to-end decision-making support technologies in different territorial management tasks is shown. Identification of environmental and anthroposphere parameters by using the developed regulatory values is the basis for early response to hazards. Risk assessment using comprehensive data analysis is used to justify priority measures to improve the safety of the territories. The use of OLAP, Data Mining, Machine Learning to jointly processing Big Data of monitoring, hazard event catalogs and territory features allows quickly and effectively preventing threats to people life.
The growing number and scale of various danger manifestations, accumulation of
large amounts of data, and development of intelligent information processing
technologies necessitate the search for new methods to support the management of
natural and technogenic territorial security. The use of analytical tools makes it
possible to effectively solve complex and multifaceted problems of managing the
integrated territorial security, to predict emergency situations (ES) of particular types.
Sufficiently good scientific elaboration of analytical methods for assessing the
territory characteristics as complex systems is often poorly implemented in practice.
The main reason is the scarcity or inaccessibility of structured data. The data analysis
methods used nowadays allow obtaining generalized background safety assessments
[
The paper presents a brief description of the developed approaches and promising directions for solving management problems using analytical technologies.
The paper is organized as follows. Section 2 reviews features of data preprocessing. Section 3 presents a description of the on-line analytical processing for integrated monitoring data. Analytical Methods for Assessing the Territorial Risks of Emergencies are discussed in Sect. 4. Finally, the conclusion is given in Section 5. 2
The use of analytical methods in the management of territories, in contrast to that of
enterprises and corporations, is in its infancy [
The studies show that most of the data that ensure solving the problems of
prevention and elimination of emergencies in Russia require the development of new
approaches to data structuring [
To fill the gaps in the generated arrays of formalized data, a distributed data
collection system was developed using the information model [
The basis for structuring monitoring data is the ontology of the subject area. The
methodology for building ontology is described in [
The advantage of the approach is the co-processing of monitoring data, hazardous event catalogs and territory features. This allows a complete description of the state of security of the territory. Frequently occurring events, such as mining-related calls, mercury spills, etc. are difficult to monitor by usual means. The ontology allows us to determine the structure and content of analytical models at the conceptual level, considering the tasks of management.
An additional source of data is mapping information used both for visualization and modeling. For example, spatial analysis can identify areas of pipelines with an increased risk of accidents based on relief models, or assess the characteristics of rivers, roads, and protective belts as obstacles for the spread of natural fires.
Providing instant access to data is especially important in the event of an emergency when it is required to formulate solutions in the shortest possible time. For example, the automatic location and identification of the phone owner can reduce the critical call time of emergency services. A request for the actual characteristics of buildings in the process of modeling the consequences of an emergency is necessary to clarify the consequences of the impact of hazardous factors on the protected objects and to form groups of forces and means corresponding to the scale of the event.
The full implementation of the structure of information resources described in the
ontology requires the creation of standardized API services for monitoring data
sources and their meta descriptions. The development of data exchange protocols is
relevant, for example, the international standards SDMX and OGC [
One of the simplest but most important analytical tasks is identifying hazards and
threats in monitoring data streams. The detection of events related to the parameters
of environmental conditions and technosphere objects exceeding the normative values
allows one to promptly notify the emergency services and inform the population. A
list of parametric and logical criteria for identifying pre-emergency and hazardous
states of the controlled objects, processes and systems is proposed in [
The joint analytical processing of the parameters for monitoring hazards and object
characteristics in the protected areas allows evaluating the possible consequences of
signaling dangers and threats. A comprehensive hazard assessment is implemented
using the OLAP technology. For example, abnormal weather events such as
extremely low temperatures, squally winds, and prolonged precipitation do not harm
infrastructure and facilities located in the Arctic zone but pose a great threat in
southern latitudes. Operational estimates of the hazards magnitude with the same
parameters can also change, depending on the capabilities of the rescue services. The
preliminary ranking of territories according to the risk degree makes it possible to
justify different criteria of hazards, depending on the location of the observation
point, frequency of the controlled events and other conditions [
Information support for management decisions in the conditions of uncertainty is
formalized in the form of information processes, including analytical modeling [
Operational analytical data processing is shown in Figure 1. Hazard identification is carried out in several stages. First, the received data packets from an external source are verified for false signals. Then, the identification of hazards is performed based on a single parameter. The regression analysis allows predicting 3-5 values, and on its basis warnings about threats are made. Analytical modules that process data from different sources work independently. This allows one to customize them for a specific data update schedule (from minutes to days), to match the expected values to the base of text templates of informational messages. Multivariate analysis is used to analyze the types of situations depending on weather, for example, flood dangers, occurrence and development of wildfires, and traffic situation.
A website and a mobile application for displaying the results of the analytical processing, as well as a service for information messages were developed. To improve the information perception, a dynamic mapping technology was developed which illustrates monitoring data in the form of special symbols and explanations for each observation point. Multivariate results are visualized using infographic libraries. The main advantage of the approach is a quick transition from the aggregated characteristics of a situation (danger signals and threats) to individual detailed parameters and their dynamics.
The implementation of this analytical processing made it possible to control the environmental parameters throughout the region, as well as those of pilot industrial facilities and communication in industrial centers. The problems identified during the operation which are related to the quality of data and the presentation of the processing results were considered in the projects of the system development for integrated operational emergency situations monitoring. 4
The priority area of using analytical methods is the assessment of territorial risks to justify strategic measures to prevent man-made disasters, mitigate consequences of natural disasters. The early studies of the risk constituents were based mainly on the expert judgment. Such coefficients, considering the peculiarities of territories, situation types, conditions of their occurrence, make it possible to assess the interaction of the system elements of any complexity. This requires the same type of a collection of experts’ opinions, which in a real situation is difficult to perform.
Using Big Data of complex monitoring, providing the interdepartmental
information exchange solves the problem concerning the lack of information
necessary for the risk assessment [
Analytical assessment makes it possible to justify the choice of the location and extent of measures to prevent emergencies and mitigate their consequences. The design of the models is implemented graphically using the Ishikawa cause-effect diagrams. The first level of charts describes the statistics of hazardous events; territory characteristics, including natural, demographic, socio-economic factors; data block for monitoring environmental parameters affecting the likelihood and scale of hazardous events. At the second level, the detailed factors are represented by independent indicators, which allow building analytical data cubes. Assessment of the state of territory security is made by comparing the indicators or calculating the integral values. 5
Using analytical processing technologies in combination with services for the consolidation of complex monitoring data has increased the efficiency of the territorial security management. The efficiency and reliability of the data used for decision-making have qualitatively changed, and the possibilities for informing the population have been expanded. Early warning about dangers and threats allows saving resources by eliminating emergencies at the inception stage, and timely help to casualties reduces losses from accidents.
Decision support implemented using the analytical technologies and big data,
provides the possibility for self-learning in the situations that have already occurred
[