KR102561776B1 - auto damage grade rating apparatus for chemical leakage accident - Google Patents

Abstract

The present invention provides an automatic damage grade calculation apparatus for a chemical leakage accident which objectively and quickly determines the scale of damage of a damaged object to quickly take follow-up action or the like. The automatic damage grade calculation apparatus comprises: an individual chemical exposure calculation module calculating individual chemical exposure information on accident area individuals from chemical leakage information of an accident area and individual location information of the accident area; an individual chemical exposure information DB; an individual risk generation module assigning individual risk values defined by the harm of leaked chemical materials and individual chemical exposure to the accident area individuals from harm information of the leaked chemical materials and the individual chemical exposure information; an individual risk DB storing the individual risk values; a risk distribution function calculation module using existing individual risk value data stored in the individual risk DB as samples to calculate a distribution function capable of calculating the distribution of individual risk values corresponding to the population of the samples and the probability of the number of people corresponding to the distribution through statistical inferences from information on the number of people of the sample; and an individual damage grade calculation module dividing the risk value distribution appearing in the risk distribution function into a plurality of sections to assign a damage grade to each section, and assigning a new individual risk value newly generated by the individual risk generation module to one among the sections to calculate an individual damage grade.

Description

Automatic damage grade rating apparatus for chemical leakage accident}

The present invention is a technology related to the post-processing of a chemical leakage accident, and more particularly, relates to an automatic damage grade calculation device for a chemical leakage accident that calculates a damage grade in a more objective manner for a target subject to a chemical leakage accident. .

Toxic chemicals are used in various industries. Regarding such chemicals, legal regulations are strengthened to ensure safer handling, but unexpected accidents may occur for various reasons.

For example, in the case of chemicals that can diffuse in the gas phase, even partial damage to the storage tank can easily leak, and leaking can cause damage to a wide area. Therefore, conventional techniques for responding to chemical leakage accidents have also been developed (eg, Republic of Korea Patent Publication No. 10-2016-0116675, etc.).

However, even if the accident response or management is carried out, the problem of compensation for damage or recovery of damage in the accident area still remains. Matters related to compensation for damage are mostly carried out according to the damage grade calculated for the target of damage (eg, individual or local government) in the accident area.

However, in the conventional damage rating calculation, there were the following problems. For example, in the case of an individual, the grade is often evaluated by evaluating the severity of the disease or other symptoms related to the leakage accident, but the evaluation items are difficult to objectify and the grade calculation is not quick, so the follow-up process is delayed. etc. are also appearing. A suitable alternative to this has not yet been proposed.

Korean Patent Publication No. 10-2016-0116675, (2016. 10. 10), specification

The technical problem of the present invention is to solve this problem, to provide an automatic damage rating calculation device for chemical leakage accidents that automatically calculates the damage rating in a more objective manner for the damage target in the event of a chemical leakage accident In this way, objectively and promptly determine the scale of damage to the victim so that follow-up measures can be taken promptly.

The technical problem of the present invention is not limited to the above-mentioned problems, and other technical problems not mentioned will be clearly understood by those skilled in the art from the following description.

The automatic damage rating calculation device for a chemical leakage accident according to the present invention is an automatic damage rating calculation device for a chemical leakage accident that automatically calculates the damage rating of a damage target from data collected in an accident area where a chemical leakage accident occurred. , Personal location information and chemical leakage information are received from the personal location information DB where personal location information of the accident area is stored and the chemical substance leakage information DB where chemical substance leakage information in the accident area is stored, respectively, and the personal chemical information for the individual in the accident area A personal chemical exposure calculation module for calculating substance exposure information; a personal chemical substance exposure information DB for storing the personal chemical substance exposure information; Personal chemical substance exposure information and hazard information of leaked chemicals are input from the personal chemical substance exposure information DB and the chemical substance hazard DB storing hazard information by chemical substance, respectively, and the personal chemical substance exposure and hazard information of leaked chemicals are input a personal risk generation module that assigns a personal risk value defined by the accident area to an individual; a personal risk DB for storing the personal risk value; Using the existing personal risk value data previously stored in the personal risk DB as a sample, the distribution of personal risk values corresponding to the population of the sample and the corresponding number of people through statistical inference from information on the number of people in the sample Risk distribution function calculation module for calculating a distribution function capable of calculating the probability of; and dividing the personal risk value distribution shown in the risk distribution function into a plurality of sections, assigning a damage grade to each section, and assigning a new personal risk value newly generated in the personal risk generation module to any one of the sections. It includes an individual damage rating calculation module that allocates and calculates individual damage ratings.

The risk distribution function may be generated as a probability density function of a continuous probability distribution in which the individual risk value is a continuous random variable.

The risk distribution function calculation module may determine the estimated fitness of the risk distribution function through a goodness-of-fit test for different types of continuous probability distributions, and the method of the goodness-of-fit test may include a maximum likelihood estimation method.

The different types of continuous probability distributions may include at least one selected from a Pareto distribution, an exponential distribution, a logistic distribution, a normal distribution, a beta distribution, a gamma distribution, and a Weibull distribution.

The personal risk generating module may store the new personal risk value in the personal risk DB after calculating the individual damage rating, and update the personal risk DB.

The automatic damage grade calculation device classifies the personal risk value data stored in the personal risk DB by accident region to generate individual risk value data by region, and the personal risk value included in the individual risk value data by region It may further include a regional risk generation module that adds up the degree values and assigns them as regional risk values for the accident area.

The automatic damage rating calculation device includes a regional risk DB for storing the regional risk value; The existing regional risk value data previously stored in the regional risk DB is used as a sample, and the distribution of regional risk values corresponding to the population of the sample and the corresponding accident are made through statistical inference from information on the number of accident regions in the sample. A regional risk distribution function calculation module for calculating a regional risk distribution function capable of calculating the probability of the number of regions; and dividing the regional risk value distribution shown in the regional risk distribution function into a plurality of intervals, assigning a regional damage grade to each interval, and assigning a new regional risk value newly generated in the regional risk generation module to any one of the intervals. A regional damage rating calculation module for calculating a regional damage rating by assigning one module may be further included.

The regional risk generation module adds up the individual risk value data for each region for each accident region, and adds a weight proportional to the size of the individual risk value included in the individual risk value data for each region. is additionally assigned to the accident area, the regional risk DB additionally stores the regional weighted risk value, and the regional risk distribution function calculation module calculates the existing regional weighted risk pre-stored in the regional risk DB. Area weighted risk that can calculate the distribution of regional weighted risk values corresponding to the population of the sample and the probability of the number of accident areas corresponding to the population of the sample through statistical inference from information on the number of accident areas in the sample using degree value data as a sample The degree distribution function is calculated, and the regional damage rating calculation module divides the regional weighted risk value distribution shown in the regional weighted risk distribution function into a plurality of sections, assigns a regional weighted damage rating to each section, and A weighted damage grade for each region can be calculated by assigning a new regional weighted risk value newly generated in the map generation module to any one of the above sections.

The regional damage rating calculation module may further calculate and provide a comprehensive damage rating for each region defined by the regional damage rating and the weighted damage rating for each region.

According to the present invention, it is possible to objectify the damage grade assessment for the victim of a chemical leakage accident and to proceed quickly in an automated manner. In particular, it is possible to calculate the damage grade more efficiently with an objectified mechanical data processing method through statistical judgment and prediction of a sample of the accident area. Therefore, follow-up processing related to damage compensation and restoration of damaged areas after an accident can be carried out quickly, so that chemical leakage accidents can be dealt with more effectively.

1 is a block diagram of an automatic damage rating calculation system for a chemical leakage accident according to an embodiment of the present invention.
FIG. 2 is a diagram conceptually illustrating information considered when generating an individual risk value generated from the automatic damage rating calculation device of FIG. 1 .
FIG. 3 is a conceptual diagram illustrating a data processing process of obtaining a risk distribution function from individual risk values in the risk distribution function calculation module of FIG. 1 .
FIG. 4 is a diagram illustrating a data processing process of calculating individual damage ratings from the individual risk value distribution of the risk distribution function in the individual damage rating calculation module of FIG. 1 .
5 is a diagram illustrating a method for calculating regional risk values of a specific accident area from individual risk values in the regional risk generation module of FIG. 1 .
6 is a conceptual diagram illustrating a data processing process for obtaining a regional risk distribution function from regional risk values in the regional risk distribution function calculation module of FIG. 1 .
FIG. 7 is a diagram illustrating a data processing process of calculating a damage grade by region from a regional risk value distribution of a regional risk distribution function in the damage rating calculation module of FIG. 1 .
8 is a diagram illustrating a method of calculating a regional weighted risk value of a specific accident area from an individual risk value in the regional risk generation module of FIG. 1 .
9 is a diagram illustrating a data processing process of calculating weighted damage grades for each region from a distribution of weighted risk values for regions calculated from weighted risk values for regions in the damage rank calculation module for each region in FIG. 1 .
10 is a diagram illustrating a method of calculating a comprehensive damage rating for each region in the damage rating calculation module for each region of FIG. 1 .

Advantages and features of the present invention and methods for achieving them will become clear with reference to the detailed description of the following embodiments in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various different forms, but only the present embodiments make the disclosure of the present invention complete and the common knowledge in the art to which the present invention belongs. It is provided to fully inform the possessor of the scope of the invention, and the invention is defined only by the claims. Like reference numerals designate like elements throughout the specification.

In this specification, 'DB' is an abbreviation for 'Data Base', and DB has the same meaning as a database. Therefore, for example, the personal risk DB means 'personal risk database'.

Hereinafter, with reference to FIGS. 1 to 10, an apparatus for automatically calculating the damage grade of a chemical leakage accident according to the present invention will be described in detail.

1 is a block diagram of an automatic damage rating calculation system for a chemical leakage accident according to an embodiment of the present invention.

Referring to FIG. 1, the automatic damage rating calculation device 1 (hereinafter referred to as the automatic damage rating calculation device) of the chemical leakage accident of the present invention processes data collected from the accident area where the chemical leakage accident occurred to determine the damage rating. automatically calculate The automatic damage rating calculation device 1 of the present invention includes a plurality of modules for data processing and calculates the damage rating of the damage target [individual and region (eg, local government)] through mechanical processing of data by the module. The objectivity of the calculation process is secured.

In addition, since the damage grade is automatically calculated and assigned through data processing, the damage grade calculation is quick. As a result, the calculation of the damage grade, which was delayed due to the conventional survey method, and the follow-up treatment (damage compensation) of the accident area after the calculation of the damage grade can be performed more quickly.

The present invention can be configured as a computer device, and each module shown in FIG. 1 can be formed by including a program capable of each calculation and/or data processing described later. For example, each of the modules shown in FIG. 1 [personal chemical exposure calculation module 100, personal risk generation module 300, risk distribution function calculation module 500, individual damage rating calculation module 600, The regional risk generation module 700, the regional risk distribution function calculation module 900, and the regional damage rating calculation module 1000] may be formed by storing a corresponding program in a computer readable recording medium. Each module may include one or more programs in combination. The recording medium may be included in a computer device, and thus each module can be driven on a computer.

In addition, each database shown in FIG. 1 (eg, personal chemical exposure information DB 200, personal risk DB 400, regional risk DB 800, and personal location information DB 10), chemical leakage information DB (20), hazard DB (30) for each chemical substance] may be formed of a mass storage device capable of inputting and recording data. Among these databases, databases that may be external to the device [eg, personal location information DB (10), chemical substance leakage information DB (20), and hazard DB (30) by chemical substance] can exchange data on the web, for example. It may also be formed in a form including a server and the like. Within these limits, the present invention can be implemented in various forms as needed.

The configuration of the automatic damage rating calculation device 1 of the present invention is as follows. The automatic damage rating calculation device (1) is an automatic damage rating calculation device (1) for a chemical leakage accident that automatically calculates the damage rating of a damaged target from data collected in an accident area where a chemical leakage accident occurred. Personal location information (S1) and chemical leakage information (S2) are input from the personal location information DB (10) in which personal location information is stored and the chemical substance leakage information DB (20) in which chemical substance leakage information in the accident area is stored. A personal chemical exposure calculation module 100 for receiving and calculating personal chemical exposure information (A) for an individual in the accident area;

Personal chemical exposure information DB (200) for storing personal chemical exposure information (A);

Personal chemical exposure information (A) and hazard information (S3) of leaked chemicals are inputted from the personal chemical exposure information DB (200) and the chemical hazard DB (30) storing hazard information by chemical substance, respectively. A personal risk generation module 300 that assigns a personal risk value (B) defined by the amount of personal chemical exposure and the hazards of the leaked chemical to the individual in the accident area,

A personal risk DB (400) for storing personal risk values (B);

Using the existing personal risk value data previously stored in the personal risk DB (400) as a sample, the distribution of personal risk values corresponding to the population of the sample and the corresponding A risk distribution function calculation module 500 for calculating a risk distribution function (C) capable of calculating the probability of the number of people, and

The personal risk value distribution shown in the risk distribution function is divided into multiple sections, a damage grade is assigned to each section, and the new personal risk value newly generated in the personal risk generation module 300 is assigned to one of the sections. and an individual damage rating calculation module 600 that allocates and calculates an individual damage rating (D).

In addition, according to an embodiment of the present invention, the automatic damage rating calculation device 1 classifies the personal risk value data stored in the personal risk DB 400 for each accident area, and the personal risk value data for each area A regional risk generation module 700 that generates a regional risk value (B-1) for the accident area by adding up the individual risk values included in the individual risk value data for each region,

A regional risk DB (800) for storing regional risk values (B-1);

Using the existing regional risk value data previously stored in the regional risk DB (800) as a sample, the distribution of regional risk values corresponding to the population of the sample and its corresponding response through statistical inference from the information on the number of accident regions in the sample. A regional risk distribution function calculation module 900 that calculates a regional risk distribution function (C-1) capable of calculating the probability of the number of accident areas in which the accident occurs, and

The regional risk value distribution shown in the regional risk distribution function is divided into a plurality of intervals, and a regional damage grade is assigned to each interval, and the new regional risk value newly generated in the regional risk generation module 700 is selected from any of the above intervals. It may further include a regional damage rating calculation module 1000 that allocates one and calculates a regional damage rating (D-1).

Through this configuration, the automatic damage rating calculation device 1 of the present invention can not only automatically calculate the individual damage rating (D) for individual victims in the accident area, but also integrate the entire accident area through appropriate data processing. For the accident area calculated as a unit area of , the damage grade (D-1) for each area can also be automatically calculated. In addition, when calculating the damage grade (D-1) for each region, as described later, the weighted damage grade (D-2) for each region is additionally calculated to determine the damage grade (D-1) and the weighted damage grade (D-2) for each region. By providing a comprehensive damage rating (D') for each region considering all of them, it is possible to objectively and quickly determine the more accurate damage scale for the accident area.

Hereinafter, the configuration and operational effects of the automatic damage rating calculation device 1 based on an embodiment of the present invention will be described in more detail. The description is based on the configuration diagram of FIG. 1 and proceeds with reference to other drawings as necessary.

Referring to FIG. 1, the automatic damage rating calculator 1 is composed of a plurality of modules for processing data and a DB for storing the data calculated by the modules, that is, a database. Each module receives data input and generates output data through independent calculation or data processing process, and output data is stored in each database and has a logical correlation in which it is calculated again in the next module. 1 conceptually shows the arrangement of each module according to the logical correlation of data processing.

The personal chemical exposure calculation module 100 calculates personal chemical exposure information (A) for an individual in an accident area. The personal chemical exposure amount calculation module 100 may include a program capable of calculating and processing data related thereto. The personal chemical exposure calculation module 100 receives personal location information (S1) and chemical leakage information (S2) of the accident area, and calculates personal chemical exposure information (A) for the individual in the accident area. . Individual chemical exposure information (A) indicates how much the individual in the accident area was exposed to the leaked chemical.

Personal chemical exposure information (A) can be calculated from personal location information (S1) of an individual in the accident area and chemical leakage information (S2) for the accident area. For example, it reflects information on where individuals were located in the accident area, where they moved around during a chemical spill, and how the chemical spread and spread throughout the accident area. Therefore, it is possible to convert an individual's exposure to the leaked chemical into an exposure amount.

For example, the personal chemical exposure information (A) is the distribution of chemical substances leaked by hour in the accident area obtained from the chemical leakage information (S2), the location of the individual in the accident area obtained from the personal location information (S1), It can be calculated as the amount multiplied by the movement path and the exposure time based on it. However, since this is just an example, it does not need to be limited in this way, and calculates individual chemical exposure information (A) by properly defining the amount that can reflect both the location information of an individual in the accident area and the chemical leakage information in the accident area. It is possible.

The personal location information DB 10 providing personal location information and the chemical leak information DB 20 providing chemical substance leak information may be outside the device, for example. For example, personal location information may include an individual's mobile phone communication record stored in a communication company, etc., and this may be provided from a communication company server or a National Police Agency server that acquires and stores it in case of an accident. Accordingly, the personal location information DB 10 may include, for example, a telecommunication company server and/or a National Police Agency server that stores location information of individuals within an accident area. However, since this is also an example, if personal location information can be obtained through another route, another server or the like can become the personal location information DB 10. The personal location information DB 10 does not need to be particularly limited to the extent that it can provide personal location information within the accident area.

The chemical leakage information may be generated including, for example, weather information of an accident area provided by the Korea Meteorological Administration. That is, it is possible to receive the diffusion form and distribution of leaked chemicals in a map form from information such as wind direction and wind speed that fluctuate during chemical leakage time within the accident area. Therefore, the chemical substance leakage information DB 20 may include the Korea Meteorological Administration server capable of providing the information, and may include a server of a research institute in which a simulator capable of simulating the degree of diffusion of chemicals according to atmospheric distribution is disposed. . However, since this is also an example, the chemical substance leakage information DB 20 does not need to be particularly limited as long as the distribution of the leaked chemical substance can be provided from weather information.

The personal chemical exposure information DB 200 stores personal chemical exposure information (A) for each individual in the accident area calculated by the personal chemical exposure calculation module 100 as described above.

The personal chemical exposure information DB 200 may generate and store personal chemical exposure information for each accident area for a plurality of chemical leakage accidents. That is, whenever a chemical leakage accident occurs or for a conventional chemical leakage accident, as described above, personal location information and chemical leakage information can be input and personal chemical exposure information can be calculated, and by continuing this, as many accidents as possible A plurality of individual chemical substance exposure information (A) calculated for a region may be stored in the personal chemical substance exposure information DB (200).

As will be described later, from the personal chemical exposure information DB 200, it is possible to receive existing and/or current personal chemical exposure information for a plurality of accident areas and perform various calculations.

The personal risk generation module 300 provides personal chemical exposure information (A) from the personal chemical exposure information DB (200) and the chemical hazard DB (30) in which hazard information for each chemical substance is stored, and leakage chemical information. Substance hazard information (S3) is input, personal chemical substance exposure amount (i.e., the contents of personal chemical substance exposure information (A)), and hazards of leaked chemicals (i.e., accidents provided from the hazard DB 30 for each chemical substance) Hazardousness of chemicals leaked into the area] to assign an individual risk value (B) to the individual in the accident area. The personal risk generation module 300 may also include a program capable of calculating and processing data related thereto.

The hazard of each chemical substance and the personal risk value (B) generated in the personal risk generation module 300 are distinguished from each other as follows. Hazardousness of each chemical substance may be, for example, a classification of chemical substances in circulation using toxicity as an index. Hazard can be, for example, a numerical value given to the degree of harm to humans according to the degree of toxicity. For example, the hazards of each chemical substance may correspond to the hazards of the Act on Registration and Evaluation of Chemicals, and in that case, it may be assigned a large value according to the degree of damage or the possibility of damage in consideration of the hazard. In such a case, the hazard DB 30 for each chemical substance may also include, for example, a server of the chemical substance information system of the National Institute of Environmental Research.

However, the hazards of each chemical do not necessarily need to be limited in this way, and other calculation methods, if necessary, [e.g., possible human exposure models in case of chemical leakage, exposure coefficients calculated by appropriate formulas, algorithms selected according to exposure scenarios It can be calculated by reflecting (calculation), etc.]. Even in such a case, the degree to which a chemical substance can harm a person can be assigned as a numerical value, and the hazard of each chemical substance can be assigned to have a large value according to the degree of damage or the possibility of damage. In this case, the hazard DB 30 for each chemical substance may also include a server of another institution that classifies and stores hazards calculated in a different way for a plurality of chemicals.

On the other hand, the individual risk value (B) is calculated by considering both the individual chemical exposure calculated for individuals in the accident area and the hazards of leaking chemicals in the accident area provided in the hazard DB 30 for each chemical. It may be a newly defined and assigned value for an individual in the accident area. For example, the individual risk value (B) may be calculated as the product of a numerically calculated personal chemical exposure of an individual at the accident area and the hazard of the leaked chemical at the accident area, which is also provided as a numerical value. The generation method of such a personal risk value (B) is conceptually shown in FIG. 2 .

FIG. 2 is a diagram conceptually illustrating information considered when generating an individual risk value generated from the automatic damage rating calculation device of FIG. 1 .

Referring to FIG. 2 , information required for a series of data processing to calculate the individual risk value (B) can be confirmed. The individual risk value (B) is the personal location information (S1) of the accident area, the personal chemical exposure information (A) calculated from the chemical leakage information (S2), and the hazard information for the leaked chemical substance provided separately. It is generated by reflecting all of (S3). In other words, the individual risk value (B) reflects how much the individual was exposed to the leaked chemical from the personal chemical exposure information (A), and at the same time, how harmful the leaked chemical is to humans. It is reflected from the hazard information (S3) of the substance. Since all of these can be expressed as numerical values, the degree of damage that can be suffered by an individual in the accident area can be calculated as an individual risk value (B) by calculating the numerical value through multiplication, etc.

However, since the individual risk value (B) may reflect the individual chemical exposure in the accident area and the harmfulness of the leaked chemical in the accident area through other formulas, it does not need to be so limited. For example, it is possible to assign a weight according to the situation, and it is also possible to define the individual risk value (B) by applying a formula other than multiplication. In any case, since it reflects both the amount of individual exposure to chemicals in the accident area and the hazards of the leaked chemical itself (hazardousness of the leaked chemical), the individual risk value that more accurately reflects the scale of individual damage in the accident area ( B) can be calculated.

In addition, since the individual risk value (B) is calculated mechanically by a predefined formula from objective information obtained from various databases, it can function as a highly objectified index for the scale of personal damage. In addition, if only data is provided, the individual risk value (B) is automatically calculated by calculation and data processing of each module (eg, personal chemical exposure calculation module 100 and personal risk generation module 300), so the accident area I can also quickly figure out the scale of my personal damage.

The personal risk value (B) calculated in this way is, for example, as the personal chemical exposure amount shown in the personal chemical exposure information (A) is larger, from the hazard information (S3) of the leaked chemical substance leaked to the accident area. The higher the hazard of the leaked chemical shown, the higher the value can be calculated and assigned to each individual in the accident area.

In the case of two or more chemical leaks in the accident area, it is possible to assign more than one individual risk value (B) to each individual in the accident area.

The personal risk DB 400 stores the personal risk value (B) generated and assigned to each individual in the accident area. Since the individual risk value (B) can also be calculated for each chemical leakage accident or for each of a number of conventional chemical leakage accidents, the personal risk DB (400) includes A number of individual risk values can be stored. In this way, through statistical data processing of individual risk values (B) generated for individuals in the accident area, it is possible to calculate the damage grade for each individual in a more objective and effective manner as described later.

Referring to FIG. 1, the risk distribution function calculation module 500 calculates the risk distribution function (C) through statistical inference from existing personal risk values previously stored in the personal risk DB 400 [FIG. 3 see (c) of]. The risk distribution function is, for example, the individual risk value corresponding to the population using the data of existing individual risk values previously stored in the personal risk DB 400 as a sample [see (a) in FIG. 3] It can be calculated as a probability distribution of

That is, the risk distribution function calculation module 500 takes existing personal risk value data previously stored in the personal risk DB 400 as a sample, and through statistical inference from information on the number of people in the sample, A distribution function capable of calculating the distribution of the corresponding individual risk value and the probability of the corresponding number of people can be calculated. In particular, the risk distribution function calculation module 500 may include a statistical data processing program capable of statistical inference, and may include a program capable of performing calculations and data processing related thereto.

FIG. 3 is a conceptual diagram illustrating a data processing process of obtaining a risk distribution function from individual risk values in the risk distribution function calculation module of FIG. 1 .

3 illustrates the distribution of existing personal risk value data pre-stored in the personal risk DB 400 [FIG. 3(a)] and the risk distribution function calculated through statistical inference [FIG. 3(c)] therefrom. (The figures in the graph are only examples). 3(b) may show an intermediate step of statistical inference (eg, generating a probability distribution object through a statistical data processing program and estimating parameters by fitting sample data). Although the calculation process of the risk distribution function is described through the graphs of FIG. 3, the graphs of FIG. 3 are only exemplified for explanation, and the present invention does not need to be limited to the drawings.

Figure 3 (a) illustrates the existing personal risk value data (horizontal axis) of a specific accident area previously stored in the personal risk DB (400) and the frequency (vertical axis) of each individual risk value. Since individual risk values are assigned to each individual in an accident area, the frequency of occurrence of a specific individual risk value is equal to the number of people (vertical axis) with the corresponding individual risk value. For example, if there are many people with low individual risk values because the amount of chemical leakage in the accident area is not large or the diffusion occurs asymmetrically, sample data in the form of FIG. 3 (a) may be provided.

Risks that can calculate the distribution of individual risk values corresponding to the population of the sample and the probability of the corresponding number of people through statistical inference from the information on the number of people in the sample using these existing individual risk value data as a sample A degree distribution function [FIG. 3 (c)] can be calculated. The risk distribution function can be calculated as, for example, a probability distribution with individual risk values as continuous random variables. Figure 3 (c) is an example of the risk distribution function calculated using Figure 3 (a) as a sample. The risk distribution function is generated as a probability density function of a continuous probability distribution with individual risk values as continuous random variables It can be.

In such a case, the vertical axis of FIG. 3 (c) may mean the probability density of the frequency (ie, the number of people), and the probability for the interval can be obtained by integrating the risk distribution function over the corresponding interval. That is, it is possible to calculate the (existence) probability of the number of people appearing in a specific section among the individual risk value distribution of the population through integration of the risk distribution function over a specific section.

The risk distribution function [Fig. 3 (c)] estimates the probability distribution of individual risk values corresponding to the population of the sample through statistical estimation using the existing individual risk value data as in Fig. 3 (a) as a sample. can be calculated in this way. For example, the risk distribution function calculation module 500 can determine the estimated goodness of fit of the risk distribution function through goodness of fit tests for different types of continuous probability distributions, and the goodness of fit test method is the maximum likelihood estimation method. estimators).

At this time, the different types of continuous probability distributions that the risk distribution function calculation module 500 can estimate as the risk distribution function include, for example, a Pareto distribution, an exponential distribution, a logistic distribution, a normal ( normal distribution, beta distribution, gamma distribution, and/or Weibull distribution. Among these distribution functions, a distribution with relatively high concordance with the existing individual risk value data can be estimated as a probability distribution for the population, and a more meaningful population probability distribution can be calculated by testing the fit through a goodness of fit test.

When estimating the type of risk distribution function, it is also possible to use, for example, a method of fitting an estimation function to a sample by generating an object on a graph as shown in FIG. 3 (b). For example, a method of generating a probability distribution object through a statistical data processing program included in the risk distribution function calculation module 500 and estimating parameters by fitting the sample data may be utilized. In addition, the risk distribution function can be calculated using various methods that can calculate the probability distribution for the population through statistical guessing.

The risk distribution function [FIG. 3 (c)] is a statistical estimate of the individual risk value distribution corresponding to the population using the existing individual risk value data previously stored as a sample, for example, as shown in FIG. 3 (a). The lower limit (a) and upper limit (b) of individual risk values (horizontal axis) shown in the risk distribution function can be increased by reflecting the size of the population. In addition, the probability distribution form of the risk distribution function may differ from the sample in terms of the peak or width of the function due to parameter selection. The risk distribution function illustrated in FIG. 3 may be a distribution of individual risk values corresponding to the population of the sample and their frequencies calculated in the form of a Pareto distribution through a process such as a goodness of fit test.

By calculating the risk distribution function in this way, it is possible to calculate the continuous distribution of individual risk values for the population using the existing individual risk value data as a sample, and the individual risk value distribution shown in the risk distribution function (horizontal axis ) into multiple sections, it is possible to more objectively calculate the individual damage rating.

Referring to FIG. 1, the individual damage rating calculation module 600 divides the individual risk value distribution shown in the risk distribution function (C) into a plurality of sections, assigns a damage rating to each section, and personal risk generation module In step 300, the newly created individual risk value is assigned to one of the divided sections to calculate the individual damage grade (D). The individual damage rating calculation module 600 may also include a program capable of calculating and processing data related thereto.

FIG. 4 is a diagram illustrating a data processing process of calculating individual damage ratings from the individual risk value distribution of the risk distribution function in the individual damage rating calculation module of FIG. 1 .

Referring to FIG. 4, the individual damage rating calculation module 600 divides the individual risk value distribution (horizontal axis) shown in the above-described risk distribution function [FIG. 3 (c)] into a plurality of sections, and the damage rating for each section can be processed to give At this time, each section may be divided so that, for example, the probability for each section is distributed by 10%. For example, each section of the damage grade shown in FIG. 4 may be set such that the probability of the risk distribution function for each section is 10%.

However, the division method of the section does not need to be limited to this, and it is possible to use other division methods as needed. It is possible to classify the distribution of individual risk values in the population by appropriate section division and assign different damage grades to each section. In addition, although 10 sections are illustrated in the drawing (that is, damage grades are divided into 10 grades), it is also possible to increase or decrease the sections as needed.

The damage grade for each individual can be selected by setting a section representing the damage grade on the risk distribution function and then matching the newly created individual risk value with a specific section. The new individual risk value can be generated in the above-described individual risk generation module 300, and is not the existing individual risk value used to generate the risk distribution function, but is newly created at the time of damage rating calculation. It can be a risk value. The new individual risk value is derived from, for example, personal location information of new accident areas newly collected in a new chemical leakage accident, chemical substance leakage information, personal chemical substance exposure information generated therefrom, and hazard information of leaked chemical substances. It may be assigned to each individual in the new accident area.

In other words, using the individual risk value data collected from the accident area as a sample, the individual risk value distribution corresponding to the population is calculated, and the calculated individual risk value distribution (the individual risk value distribution shown on the risk distribution function) ) can be divided into sections and used as a criterion for damage rating. As illustrated in FIG. 4 , the newly created personal risk value may be assigned to a specific section among the divided sections, and a damage grade may be calculated corresponding to the assigned section. For example, if a new individual risk value is matched with a section corresponding to a damage grade of 7 as shown in FIG. It can be.

The calculation of the damage grade may be performed at once for each of a plurality of individuals identified in the new accident area. That is, a plurality of new individual risk values are automatically generated for a plurality of individuals in a new accident area from the individual risk generation module 300 through the above-described data processing process, and each of these individuals appears in the risk distribution function. By matching on the divided sections of the risk value distribution, the damage rating for each corresponding individual can be calculated.

The personal risk generation module 300 may store a new personal risk value in the personal risk DB 400 after calculating the damage rating for each individual, and update the personal risk DB 400 immediately. Therefore, when the next individual damage rating is calculated, it is possible to calculate the individual damage rating by calculating a new risk distribution function from the updated personal risk DB 400 and reflecting the increased data. This process may be repeated whenever new data is input.

In this way, it is possible to more objectively and quickly calculate individual damage grades in a statistical manner using the automatic damage grade calculation device 1 of the present invention.

Meanwhile, referring to FIG. 1, the automatic damage rating calculation device 1 of the present invention includes a regional risk generation module 700, a regional risk DB 800, a regional risk distribution function calculation module 900, and a regional risk distribution function module 900. Including the damage rating calculation module 1000, it is possible to automatically calculate damage ratings for each region given to the entire accident area. The related contents will be described in more detail below.

The calculation of the regional damage grade (D-1) is conceptually similar to the above-mentioned calculation of the individual damage grade (D). However, instead of the above-mentioned individual risk value, a regional risk value is additionally generated and used as a variable to calculate the damage grade by region.

The regional risk generation module 700 classifies the personal risk value data stored in the personal risk DB 400 for each accident region, generates individual risk value data for each region, and stores the individual risk value data for each region. The included individual risk values are added up and assigned as a regional risk value (B-1) for the accident area. The regional risk generation module 700 may also include a program capable of calculating and processing related data.

5 is a diagram illustrating a method for calculating regional risk values of a specific accident area from individual risk values in the regional risk generation module of FIG. 1 .

5 illustrates a method for calculating regional risk values. The regional risk generation module 700 classifies the personal risk value data stored in the personal risk DB 400 for each accident area where each accident occurred. For example, individual risk value data for each region corresponding to each accident region from accident region 1 to accident region n (n is a natural number) may be generated.

The individual risk value data for each region classified for each accident region are added up for each region and assigned as a regional risk value for the corresponding accident region. That is, the regional risk value of accident area 1 can be calculated as the sum of all individual risk values included in the individual risk value data for each area of accident area 1. Therefore, the regional risk value of a specific accident area can be influenced by the population within the accident area and the distribution of individual risk values within the accident area. As will be described later, it is possible to comprehensively calculate the damage grade for each region by calculating the regional weighted risk value (B-2 in FIG. This will be explained again later.

Since the regional risk value is calculated as the sum of the individual risk values, the regional risk value can be generated practically in any accident area to which individual risk values are assigned. That is, in a specific accident area, while generating an individual risk value to calculate the damage rating for each individual, at the same time, the regional risk value for the entire accident area is automatically calculated through the sum of the individual risk values assigned to each individual. it is possible

As described above, individual risk values for a plurality of accident areas may be stored in the personal risk DB (400 in FIG. 1), respectively. A regional risk value can be calculated for each. The regional risk generation module 700 may generate regional risk values for a plurality of accident regions in this manner and assign them to each accident region.

Referring to FIG. 1, the regional risk DB 800 stores regional risk values (B-1) generated and assigned to each accident region. Area risk values can also be continuously updated by being assigned to the accident area whenever a new accident occurs. It can also be updated when data on previous accident areas are added.

The regional risk distribution function calculation module 900 may perform calculations in substantially the same manner as the risk distribution function calculation module 500 described above. However, while the risk distribution function calculation module 500 calculates the risk distribution function for an individual by performing statistical processing on the individual risk value, the regional risk distribution function calculation module 900 calculates the above-mentioned regional risk distribution function. Calculate the regional risk distribution function (C-1) through statistical processing of risk values.

That is, the regional risk distribution function calculation module 900 takes existing regional risk value data previously stored in the regional risk DB 800 as a sample, and statistically infers the sample from information on the number of accident regions in the sample. Calculate the regional risk distribution function that can calculate the distribution of regional risk values corresponding to the population of and the probability of the number of accident regions corresponding to it. The regional risk distribution function calculation module 900 may also include a statistical data processing program capable of statistical inference, and may be configured to include a program capable of calculating and processing data related thereto.

6 is a conceptual diagram illustrating a data processing process for obtaining a regional risk distribution function from regional risk values in the regional risk distribution function calculation module of FIG. 1 .

6 shows the distribution of existing regional risk value data previously stored in the regional risk DB 800 [Fig. 6(a)] and the regional risk distribution function calculated through statistical inference [Fig. 6(c)] Illustrated (the figures in the graph are only examples). 6(b) may show an intermediate step of statistical inference (eg, generating a probability distribution object through a statistical data processing program and estimating parameters by fitting sample data). The calculation process of the regional risk distribution function is also described through the graphs of FIG. 6, but since the graphs are only illustrated for explanation, the present invention does not need to be limited to the drawings.

6 (a) illustrates existing regional risk value data (horizontal axis) for multiple accident regions previously stored in the regional risk DB (800) and the frequency (vertical axis) of each regional risk value. Since regional risk values are assigned to each accident area, the frequency of occurrence of regional risk values is the same as the number of accident areas (vertical axis) assigned with corresponding regional risk values. For example, when low regional risk values are assigned to most accident regions, sample data in the form of FIG. 6 (a) can be provided.

Using these existing regional risk value data as a sample, through statistical inference from the information on the number of accident regions in the sample, the distribution of regional risk values corresponding to the population of the sample and the probability of the corresponding number of accident regions are calculated A possible regional risk distribution function [Fig. 6(c)] can be calculated. The regional risk distribution function can also be calculated as a probability distribution with regional risk values as continuous random variables. Figure 6 (c) is an example of the regional risk distribution function calculated using Figure 6 (a) as a sample. The regional risk distribution function is a probability density function of a continuous probability distribution with regional risk values as continuous random variables can be created with

In such a case, the vertical axis of FIG. 6 (c) can be represented by the probability density of the frequency (ie, the number of accident areas), and the probability for the interval can be obtained by integrating the regional risk distribution function over the corresponding interval. In other words, it is possible to calculate the (existence) probability of the number of accident regions for a specific section among the distribution of regional risk values of the population through integration of a specific section of the regional risk distribution function.

The regional risk distribution function [FIG. 6(c)] takes the existing regional risk value data as shown in FIG. It can be calculated in an estimated way. The regional risk distribution function calculation module 900 can determine the estimated fitness of the regional risk distribution function through a goodness-of-fit test for different types of continuous probability distributions, and the test method for goodness of fit is maximum likelihood estimators etc. may be included.

At this time, different types of continuous probability distributions that can be estimated by the regional risk distribution function calculation module 900 as regional risk distribution functions are also, for example, Pareto distribution, exponential distribution, logistic distribution, It may include a normal distribution, a beta distribution, a gamma distribution, and/or a Weibull distribution. Among these distribution functions, a distribution with relatively high concordance with the existing regional risk value data can be estimated as a probability distribution for the population, and a more meaningful probability distribution of the population can be calculated by testing the fit through a goodness-of-fit test.

When estimating the type of regional risk distribution function, it is also possible to use, for example, a method of fitting an estimation function to a sample by generating an object on a graph as shown in FIG. For example, a method of generating a probability distribution object through a statistical data processing program included in the regional risk distribution function calculation module 900 and fitting it to sample data to estimate parameters may be utilized. In addition, the regional risk distribution function can be calculated using various methods that can calculate the probability distribution for the population through statistical estimation.

The regional risk distribution function [Fig. 6 (c)] shows the regional risk value distribution corresponding to the population by using the existing regional risk value data previously stored as a sample, for example, as shown in Fig. 6 (a), so it is a regional risk The lower limit (a') and upper limit (b') of the regional risk value (horizontal axis) shown in the degree distribution function can also be increased by reflecting the size of the population. In addition, the probability distribution form of the regional risk distribution function may differ from the sample in terms of the peak or width of the function due to parameter selection. The regional risk distribution function illustrated in FIG. 6 may be a distribution of regional risk values corresponding to the population of the sample and their frequencies calculated in the form of a Pareto distribution through a process such as a goodness of fit test.

By calculating the regional risk distribution function in this way, it is possible to calculate the continuous distribution of regional risk values for the population using the existing regional risk value data as a sample, and the regional risk value distribution shown in the regional risk distribution function. It is possible to more objectively calculate the damage grade by region by dividing (horizontal axis) into multiple sections.

Referring to FIG. 1, the regional damage rating calculation module 1000 divides the regional risk value distribution shown in the regional risk distribution function (C-1) into a plurality of sections, assigns a regional damage rating to each section, and A new regional risk value newly generated in the risk generation module 700 is assigned to one of the divided sections to calculate a damage grade (D-1) for each region. The regional risk generation module 700 may include a program capable of calculating and processing data related thereto.

FIG. 7 is a diagram illustrating a data processing process of calculating a damage grade by region from a regional risk value distribution of a regional risk distribution function in the damage rating calculation module of FIG. 1 .

Referring to FIG. 7, the regional damage rating calculation module 1000 divides the regional risk value distribution (horizontal axis) shown in the above-described regional risk distribution function [Fig. 6(c)] into a plurality of sections, and each section A local damage rating can be assigned. At this time, each section may be divided so that, for example, the probability for each section is distributed by 10%. For example, each section of the regional damage rating shown in FIG. 7 may be set such that the probability of the regional risk distribution function for each section is 10%.

However, the division method of the section does not need to be limited to this, and other types of division methods can be used as needed. It is possible to divide the regional risk value distribution of the population by appropriate section division and assign different regional damage ratings to each section. In addition, although 10 sections are illustrated in the drawing (that is, local damage grades are divided into 10 grades), it is also possible to increase or decrease the sections as needed.

The damage grade for each region can be selected by setting a section representing the regional damage grade on the regional risk distribution function and then matching the newly created regional risk value with a specific section. The new regional risk value can be generated in the regional risk generation module 700, and is not the existing regional risk value used to generate the regional risk distribution function, but is newly created at the time of regional damage rating calculation. It can be a risk value. For example, the new area risk value may be assigned as the sum of all new individual risk values newly generated for the new accident area.

That is, the regional risk value distribution corresponding to the population is calculated using the regional risk value data collected from the accident area as a sample, and the calculated regional risk value distribution (regional risk value shown on the regional risk distribution function) distribution) can be divided into sections and used as a standard for regional damage ratings. As illustrated in FIG. 7 , the newly created regional risk value may be assigned to a specific section among the divided sections, and a damage grade for each region may be calculated corresponding to the assigned section. For example, if a new regional risk value is matched with a section corresponding to a regional damage rating of 5, as shown in FIG. ratings can be calculated.

The calculation of damage ratings for each region may also be performed for each of a plurality of new accident regions when there are two or more new accident regions. That is, new regional risk values for a plurality of new accident areas are generated at once from the regional risk generation module 700 through the above-described data processing process, and the regional risk value distribution shown in the regional risk distribution function for each of them. It is possible to calculate the damage grade for each corresponding region by matching on the divided section of .

The regional risk generation module 700 may also store new regional risk values in the regional risk DB 800 after calculating the damage rating for each region, and immediately update the regional risk DB 800. Therefore, next time when the damage grade for each region is calculated, it is possible to calculate the damage grade for each region by calculating a new regional risk distribution function from the updated regional risk DB 800 and reflecting the increased data. This process may also be repeated whenever new data is input.

In this way, by using the automatic damage rating calculation device 1 of the present invention, it is possible to more objectively and quickly calculate the damage rating for each region in a statistical manner.

On the other hand, referring to FIG. 1, the automatic damage rating calculation device 1 of the present invention calculates the regional weighted damage rating (D-2) that reflects the weighted damage scale by region when calculating the regional damage rating (D-1). can do. Finally, the comprehensive damage scale for the region can be identified by considering both the damage grade by region and the weighted damage grade by region. Hereinafter, this will be described in more detail.

Regional weighted damage ratings are also calculated conceptually similar to the aforementioned regional damage ratings. However, the regional weighted damage rating is calculated through statistical processing of the regional weighted risk value (B-2) instead of the regional risk value described above. The calculation of the weighted damage grade for each region can also be performed from the aforementioned regional risk generation module (700), regional risk DB (800), regional risk distribution function calculation module (900), and regional damage rating calculation module (1000). , each module can further act as follows.

Referring to FIG. 1, the regional risk generation module 700 classifies the above-described individual risk value data for each region (as described above, the individual risk value data stored in the personal risk DB 400 for each accident region). generated] for each accident area, but a regional weighted risk value (B-2) with a weight proportional to the size of the individual risk value included in the individual risk value data for each area is added for the accident area grant,

The regional risk DB 800 additionally stores regional weighted risk values,

The regional risk distribution function calculation module 900 uses the existing regional risk value data previously stored in the regional risk DB 800 as a sample through the above-described statistical processing, and makes statistical inference from information on the number of accident regions in the sample. Calculate a regional weighted risk distribution function (C-2) capable of calculating the distribution of regional weighted risk values corresponding to the population of the sample and the probability of the number of accident areas corresponding thereto,

The regional damage rating calculation module 1000 divides the regional weighted risk value distribution shown in the regional weighted risk distribution function into multiple sections, assigns a regional weighted damage rating to each section, and in the regional risk generation module 700 A weighted damage grade (D-2) for each region can be calculated by assigning a newly created regional weighted risk value to any one of the above sections.

The process of calculating weighted damage ratings by region is practically identical to the process of calculating damage ratings by region described above, except that regional weighted risk values are used as variables instead of regional risk values. Therefore, it will be described with reference to the calculation process of the damage rating for each region described above.

8 is a diagram illustrating a method for calculating a regional risk weighted risk value of a specific accident area from an individual risk value in the regional risk generation module of FIG. 1, and FIG. It is a diagram illustrating the data processing process of calculating the weighted damage grade for each region from the regional weighted risk value distribution calculated from the risk value.

Referring to FIG. 8, the regional weighted risk value is described as follows. A regional weighted risk value is also assigned for each accident area, just like the regional risk value. However, for the regional weighted risk value, a weight proportional to the size of the individual risk value included in the data is added when summing up the individual risk value data by region for each accident area.

Therefore, for example, even in the same accident area, the size of the regional weighted risk value may be larger than the regional risk value, and for different accident areas, an accident area having multiple individual risk values with a relatively large size (i.e., Accident areas with many high-risk individuals) can generate a larger size of the regional weighted risk value. For example, weights can be applied in such a way as to multiply each individual risk value that is added up to calculate a regional weighted risk value by a number proportional to the size, so the weight can also vary depending on the size of the individual risk value. . The weight may be modified and applied in various ways including the exemplified method.

In this way, each regional weighted risk value can be generated for each accident area by adding the weighted individual risk values. That is, as shown in FIG. 8, n area weighted risk values may be assigned to n accident areas. Since the aforementioned regional risk values are also assigned to each accident region, a regional risk value and a regional weighted risk value can be assigned as a pair for all accident regions with data.

If the regional weighted risk value is calculated, as described above, the regional risk distribution function can be calculated through a process substantially equivalent to the process of calculating the regional risk distribution function through statistical inference from the regional risk value. there is. That is, the regional risk distribution function can be calculated by using a statistical processing method equivalent to the regional risk distribution function calculation process of FIG. 6 described above, but replacing the variable with regional weighted risk values instead of regional risk values.

Therefore, since the calculation process of the regional risk distribution function can be understood in accordance with the calculation process and related description of the regional risk distribution function of FIG. 6 described above, a repeated description thereof will be omitted. However, even in the same accident area, the size of the regional weighted risk value is different from the regional risk value, so even if statistical processing is performed in the same way, the shape of the sample and the shape of the population are different from those of the regional risk value.

When the regional weighted risk distribution function is calculated in such a way, as shown in FIG. A grade can be assigned. The division method of the section also corresponds to the division method of the regional risk value described above [see FIG. 7 and related description], so it can be understood accordingly.

The regional weighted damage rating can be selected by matching the newly created regional weighted risk value with a specific interval after setting the interval representing the regional weighted damage rating on the regional weighted risk distribution function in this way. The new regional weighted risk value is also generated by the regional risk generation module 700, and is not the existing regional weighted risk value used to generate the regional weighted risk distribution function. It may be a generated regional weighted risk value. The new area weighted risk value may be generated by, for example, a weighted sum of all new individual risk values newly generated for the new accident area.

Therefore, the regional weighted risk value distribution corresponding to the population was calculated using the regional weighted risk value data collected from the accident area as a sample, and the calculated regional weighted risk value distribution (the area shown on the regional weighted risk distribution function) Weighted risk value distribution) can be divided into sections and used as a standard for regional weighted damage ratings. As illustrated in FIG. 9 , the newly created regional weighted risk value may be assigned to a specific section among the divided sections, and a weighted damage rating for each region may be calculated corresponding to the assigned section. For example, if the new regional weighted risk value is matched with a section corresponding to a regional weighted damage rating of 5, as shown in FIG. Weighted damage ratings for each region can be calculated by rating.

The calculation of the weighted damage rating for each region may also be performed for each of a plurality of new accident regions when there are two or more new accident regions. That is, as in the case of calculating the damage rating for each region described above, new region weighted risk values for a plurality of new accident regions are generated at once from the regional risk generation module 700, and each of them is a region represented by the regional weighted risk distribution function It is possible to calculate the weighted damage grade for each corresponding region by matching on the divided sections of the weighted risk value distribution.

The regional risk generation module 700 may store new regional weighted risk values in the regional risk DB 800 after calculating the weighted damage rating for each region, and immediately update the regional risk DB 800. Therefore, when the weighted damage grade for each region is calculated next time, it is possible to calculate the weighted damage grade for each region by calculating a new regional weighted risk distribution function from the updated regional risk DB (800) and reflecting the increased data. This process may also be repeated whenever new data is input.

Therefore, the automatic damage rating calculation device 1 of the present invention generates individual risk values for each accident area and automatically calculates the individual damage rating (D) for individuals in the accident area through statistical processing thereof. , From the regional risk value assigned to each accident area and the regional weighted risk value, a pair of regional damage ratings (D-1) and regional weighted damage ratings (D-2) were obtained for each accident region through the same statistical processing. can be calculated as

In particular, the regional damage rating calculation module 1000 calculates the damage rating and weighted damage rating representing the scale of damage by region, and further calculates and provides a comprehensive damage rating for each region defined by the regional damage rating and weighted damage rating for each region. Through this, it is possible to more accurately calculate the scale of damage to a specific accident area.

10 is a diagram illustrating a method of calculating a comprehensive damage rating for each region in the damage rating calculation module for each region of FIG. 1 .

The comprehensive damage rating for each region (D′ in FIG. 1) may be calculated as, for example, the product of the regional damage rating (D-1 in FIG. 1) and the weighted damage rating for each region (D-2 in FIG. 1). For example, referring to FIG. 10, when the damage grade by region and the weighted damage grade by region are calculated as 4 and 7 respectively in a specific accident area (eg, accident area 1) (the figures in the graph are exemplary), comprehensive damage by area The rating may be calculated as a numerical value corresponding to the product. For other areas (accident areas 2 and 3) illustrated in FIG. 10 , comprehensive damage ratings for each area may be calculated in the same manner.

Therefore, when both the regional damage rating and the regional weighted damage rating have a large value (eg, accident area 3), a relatively higher overall damage rating by region can be calculated. Moreover, as compared to accident areas 1 and 2, even if the damage rating by area is relatively low (accident area 1), if the weighted damage rating by area is higher, it is possible to calculate the overall damage rating by area as a higher value. The regional weighted damage rating reflects the above-mentioned regional weighted risk value, and the regional weighted risk value reflects the weight proportional to the size of the individual risk value of each individual in the accident area as described above. In addition to the population in the difficult accident area, the scale of individual damage can be more accurately reflected. Therefore, it is possible to more objectively and accurately grasp the scale of damage to the accident area through the calculation of the comprehensive damage rating by region considering this.

Although the embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art can realize that the present invention can be implemented in other specific forms without changing the technical spirit or essential features. you will be able to understand Therefore, the embodiments described above should be understood as illustrative in all respects and not limiting.

1: Automatic damage rating calculation system for chemical leakage accidents
10: personal location information DB 20: chemical substance leakage information DB
30: Hazard DB by chemical substance 100: Personal chemical exposure calculation module
200: Personal chemical exposure information DB 300: Personal risk generation module
400: Personal risk DB 500: Risk distribution function calculation module
600: individual damage rating calculation module 700: regional risk generation module
800: regional risk DB 900: regional risk distribution function calculation module
1000: Damage rating calculation module for each region
A: Personal chemical exposure information B: Personal risk value
B-1: Regional risk value B-2: Regional weighted risk value
C: risk distribution function C-1: regional risk distribution function
C-2: Regional weighted risk distribution function D: Individual damage rating
D-1: Damage rating by region D-2: Weighted damage rating by region
D': Comprehensive damage rating by region S1: Personal location information
S2: Chemical Leak Information S3: Hazard Information of Leaked Chemicals

Claims (9)

In the automatic damage grade calculation device for chemical leakage accident that automatically calculates the damage grade of the damage target from the data collected in the accident area where the chemical leakage accident occurred,
Personal location information and chemical leakage information are inputted from the personal location information DB in which personal location information of the accident area is stored and the chemical substance leakage information DB in which chemical substance leakage information in the accident area is stored, respectively Personal chemical exposure calculation module for calculating exposure information;
a personal chemical substance exposure information DB for storing the personal chemical substance exposure information;
Personal chemical substance exposure information and hazard information of leaked chemicals are input from the personal chemical substance exposure information DB and the chemical substance hazard DB storing hazard information by chemical substance, respectively, and the personal chemical substance exposure and hazard information of leaked chemicals are input a personal risk generation module that assigns a personal risk value defined by the accident area to an individual;
a personal risk DB for storing the personal risk value;
Using the existing personal risk value data previously stored in the personal risk DB as a sample, the distribution of personal risk values corresponding to the population of the sample and the corresponding number of people through statistical inference from information on the number of people in the sample A risk distribution function calculation module for calculating a risk distribution function capable of calculating a probability of ; and
The personal risk value distribution shown in the risk distribution function is divided into multiple sections, a damage grade is assigned to each section, and the new personal risk value newly generated in the personal risk generation module is assigned to one of the sections. It includes an individual damage rating calculation module that calculates individual damage ratings by
The individual risk value is,
The personal chemical exposure calculated as a numerical value and the harmfulness of the leaked chemical provided as a numerical value are calculated by a defined formula, and the degree of damage that can be suffered by an individual in the accident area is automatically calculated as a numerical value,
The risk distribution function,
It is generated as a probability density function of a continuous probability distribution with the individual risk value corresponding to the population of the sample as a continuous random variable,
The damage grade is
The probability of the risk distribution function is given differently for each segment divided so that it is distributed,
The damage rating for each individual is,
Automatic damage rating calculation device for chemical leakage accidents, characterized in that the calculation is calculated for each individual to whom the new individual risk value matched with the interval is assigned. delete According to claim 1,
The risk distribution function calculation module determines the estimated fitness of the risk distribution function through a goodness-of-fit test for different types of continuous probability distributions, and the chemical leakage accident in which the goodness-of-fit test method includes the maximum likelihood estimation method Automatic damage rating calculation system. According to claim 3,
The different types of continuous probability distributions include one or more selected from a Pareto distribution, an exponential distribution, a logistic distribution, a normal distribution, a beta distribution, a gamma distribution, and a Weibull distribution. According to claim 1,
The personal risk generating module, after calculating the individual damage rating, stores the new personal risk value in the personal risk DB and updates the personal risk DB. According to claim 1,
The individual risk value data stored in the personal risk DB is classified by accident area to create individual risk value data by area, and the individual risk values included in the personal risk value data by area are added up to determine the accident area. An automatic damage rating calculation device for chemical leakage accidents that further includes a regional risk generation module, which is given as a regional risk value for the region. According to claim 6,
a regional risk DB for storing the regional risk value;
The existing regional risk value data previously stored in the regional risk DB is used as a sample, and the distribution of regional risk values corresponding to the population of the sample and the corresponding accident are made through statistical inference from information on the number of accident regions in the sample. A regional risk distribution function calculation module for calculating a regional risk distribution function capable of calculating the probability of the number of regions; and
The regional risk value distribution shown in the regional risk distribution function is divided into a plurality of intervals, and a regional damage grade is assigned to each interval, and the new regional risk value newly generated in the regional risk generation module is selected as one of the intervals. Automatic damage rating calculation device for chemical leakage accidents further comprising a regional damage rating calculation module for calculating regional damage ratings by assigning to. According to claim 7,
The regional risk generation module adds up the individual risk value data for each region for each accident region, and adds a weight proportional to the size of the individual risk value included in the individual risk value data for each region. is additionally granted to the accident area,
The regional risk DB additionally stores the regional weighted risk value,
The regional risk distribution function calculation module takes existing regional risk value data previously stored in the regional risk DB as a sample, and through statistical inference from information on the number of accident regions in the sample, Calculate a regional weighted risk distribution function that can calculate the distribution of regional weighted risk values and the probability of the number of accident areas corresponding to it,
The regional damage rating calculation module divides the regional weighted risk value distribution shown in the regional weighted risk distribution function into a plurality of sections, assigns a regional weighted damage rating to each section, and newly generated in the regional risk generation module. An automatic damage rating calculation device for chemical leakage accidents that calculates a weighted damage rating for each region by assigning a new regional weighted risk value to any one of the above sections. According to claim 8,
The damage rating calculation module for each region further calculates and provides a comprehensive damage rating for each region defined by the regional damage rating and the weighted damage rating for each region.

Images