JPS61166673A - Processing system for disaster foreseeing information - Google Patents

Abstract

PURPOSE:To attain the search for various factors of the disaster generating mechanism and to improve the foreseeing accuracy, by giving the weighting to the evaluation value of the set disaster foreseeing measurement items and discriminating the related groups of said measurement items through the function processing. CONSTITUTION:A discriminating function arithmetic part 4 obtaines the discriminating function of the working site where a disaster occurs for each disaster generated site and each disaster foreseeing item by totalizing the conversion coefficients of store parts 1 and 2 and the multiplied value of evaluated values. At the same time, the discriminating function of the no-disaster generated working site group is also obtained. While an operating working site function arithmetic part 5 obtains both the disaster generated group function and the no-disaster generated working site group function for each operating working site. Then a disaster foreseeing/discriminating part 6 decides either one of both discriminating functions obtained by the part 5 to which the group is belonging through comparison. Thus it is possible to search various factors of the disaster generating mechanism and to improve the disaster foreseeing accuracy.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention provides a disaster prediction system that sets disaster prediction measurement items, calculates a function based on the evaluation value, and provides disaster prediction information for each operating workplace. It relates to information processing systems.

[Conventional technology]

However, it is significantly higher than other industries.

Work sites in the construction industry are scattered all over the place, ranging from small scale to large scale, and for each work,
Safety management is carried out according to location conditions, contents, construction methods, etc. In the disaster prediction information processing system separately proposed by the inventors of this application, for example, the location conditions and contents of construction,
The disaster prediction index is calculated as the degree of disaster risk based on the specific technology and information of the construction method, etc., and the disaster prediction index is provided to the relevant work site, so that the disaster prediction index exceeds the specified standard value. Focused safety guidance is given to workplaces. Disaster risk in this system is determined by evaluating and detecting the target items of the disaster occurrence mechanism.
From this evaluation, the disaster incidence rate (probability of occurrence based on past evaluation values) is calculated and added.

[Problem that the invention seeks to solve]

However, in the conventional system that calculates the degree of disaster risk by simple addition, when a disaster occurs due to the accumulation of various factors, the degree of risk is increased. If it occurs due to interaction, the ability to predict and detect it will be low.

The present invention solves the above problems and aims to provide a disaster prediction information processing system that can improve prediction accuracy.

[Means for solving problems]

To this end, the disaster prediction information processing system of the present invention sets disaster prediction measurement items, obtains evaluation values of the disaster prediction measurement items for each work site, and calculates the evaluation value and a predetermined constant using a function. This is a disaster prediction information processing system that provides disaster prediction information for each operating workplace based on the following: a disaster-occurring workplace group discriminant function and a no-accident workplace group discriminant function from evaluation values of disaster-occurring workplaces and non-disaster-occurring workplaces. From the discriminant function calculation means that calculates the
,,,.

η A group function calculation means for calculating a group function for each operating workplace, and provides disaster prediction information by determining the similarity between the group function and a disaster-occurring workplace group discriminant function and an accident-free workplace group discriminant function. It is characterized by this.

[Effect]

For workplaces that are operating in similar conditions to workplaces that have repeatedly caused accidents in the past (workplaces where disasters have occurred), the evaluation values of disaster prediction measurement items will be similar to those of workplaces where disasters have occurred. For the accident-occurring workplace group discriminant function and accident-free workplace group discriminant function calculated from the evaluation values of accident-occurring workplaces and accident-free workplaces, a group function obtained by similarly calculating for operating workplaces is used. By checking whether the workplace group is similar to the disaster-occurring workplace group or the accident-free workplace group, it is possible to provide prediction information about the disaster-occurring workplace with extremely high accuracy.

〔Example〕

Examples will be described below with reference to the drawings.

FIG. 1 is a diagram showing the configuration of one embodiment of the disaster prediction information processing system of the present invention. In the figure, 1 is a coefficient storage section, 2
and 3 indicate an evaluation value storage section, 4 a discriminant amount prediction discrimination determination section, and 7 an input section, respectively.

In FIG. 1, a coefficient storage unit 1 stores a constant A given to a workplace where a disaster has occurred, a conversion coefficient A= (i=1.2...) for each disaster prediction measurement item, and a non-accident operation. The evaluation value storage unit 2 stores the constant a0 given to the workplace and the conversion coefficient ai (i=1+2+...) for each disaster prediction measurement item. Evaluation value for predictive measurement item X = (
i=1.2. ...), and the evaluation value storage unit 3 stores the evaluation value xt (i=L2+...) for each disaster prediction measurement item for the operating workplace.
It is used to store. Here, the constant determines how much weight should be given to which disaster prediction measurement item (disaster occurrence factor).

The discriminant function calculation unit 4 calculates a discriminant function 21 for the accident-occurring workplace group and a discriminant function 22 for the accident-free workplace group. Here, the discriminant function 2 for the workplace group where the disaster occurred. are the conversion coefficient A for the disaster prediction measurement item of the disaster occurrence workplace stored in each disaster occurrence work, the evaluation value X for each disaster prediction measurement item of the disaster occurrence workplace stored in the evaluation value storage unit 2, and It is obtained by multiplying by , and summing them together with a constant A0. In addition, the discriminant function 22 for the accident-free workplace group includes, for each disaster prediction measurement item for each working workplace, the conversion coefficient a for the disaster prediction measurement item of the accident-free workplace stored in the coefficient storage unit 1, and the evaluation value. Evaluation value X for each disaster prediction measurement item of the operating workplace stored in the storage unit 3.

It is obtained by multiplying by and summing together with a constant a0. In other words, here, the discriminant function 22 of the accident-free workplace group is calculated by applying the working workplaces as the accident-free workplace group, but the evaluation values of the disaster prediction measurement items of past accident-free workplaces are used. It goes without saying that you may do as you like. That is, in the discriminant function calculation unit 4, Z+''Ao+ΣA,X.

Z2 = ao + Σai Xi calculations are performed.

The operating workplace function calculation unit 5 calculates a disaster occurrence workplace group function 2 for each operating workplace. and accident-free workplace group function 22. Here, the disaster occurrence workplace group function z stores the conversion coefficient A and the evaluation value for the disaster prediction measurement item of the disaster occurrence workplace stored in the coefficient storage unit 1 for each disaster prediction measurement item for each operating workplace. It is obtained by multiplying the evaluation value X for each disaster prediction measurement item of the working workplace stored in the section 3, and summing the result together with a constant A0. In addition, the accident-free workplace group function 22 includes a conversion coefficient 38 for the disaster prediction measurement item of the accident-free workplace stored in the coefficient storage unit 1 and an evaluation value storage unit 3 for each disaster prediction measurement item for each operating workplace. Evaluation value x for each disaster prediction measurement item of the operating workplace stored in
Multiply by 8 and 67 constant °°'! ″8°° 8+su2′.2
°°1°7 I□... can be found. That is, in the operating workplace function calculation unit 5, Z 1 =A a +ΣAi xizz = ao +
An operation of Σai xi is performed.

The disaster prediction discrimination judgment unit 6 determines the operating work individually for each operating workplace based on the distribution of the discriminant function ZI of the disaster-occurring workplace group and the distribution of the discriminant function 22 of the accident-free workplace group obtained by the discriminant function calculation unit 4. It compares and determines which discriminant function the disaster-occurring workplace group function 21 and the accident-free workplace group function Zt obtained by the workplace function calculation unit S are similar to, and the operating workplace belongs to the disaster-occurring workplace group. This is to determine whether the workplace belongs to the accident-free workplace group.

The human power unit 7 manually inputs data to be stored in the coefficient storage unit 1 and the evaluation value storage unit 2.3, and issues calculation commands to the discriminant function calculation unit 4, operating work place function calculation unit 5, and disaster prediction discrimination determination unit 6. It is something that you input.

Next, a specific example of disaster prediction information processing according to the present invention will be explained. Figure 2 is a diagram for explaining examples of fixed values for disaster prediction measurement items, Figure 3 is a diagram for explaining characteristics of disaster prediction measurement items and harm prediction measurement items by workplace group, and Figure 5 is a diagram for explaining the characteristics of disaster prediction measurement items and harm prediction measurement items by workplace group. It is a figure showing an example of a judgment result outputted by a disaster prediction information processing system of the present invention.

Disaster prediction measurement items can be divided into fixed values that are evaluated depending on the content of the construction work and variable values that are evaluated as the construction progresses. In other words, the fixed value is based on the characteristic factors such as construction classification, structure, total area, etc. as shown in Figure 2.
In the case of a building, the degree of difficulty of construction is evaluated based on the number of floors (underground, above ground), construction period and other factors, and the history of disasters is evaluated based on past disaster occurrence data of the person in charge of construction and the work site. There may or may not be dangerous construction processes such as steel frame erection work and scaffolding construction work. Further, as shown in FIG. 3, the variation values include, as characteristic factors, the implementation status of advance planning, the implementation status of the safe construction cycle, the evaluation of the safety management status, and the history evaluation of the correction status. These data may be generated from construction reports, construction schedules, various types of charts, or periodically manually as shown in FIGS. 2 and 3, respectively. Then, based on these evaluation values, the disaster risk distribution is calculated using numerical values calculated based on past experience. For example, the disaster risk level can be defined by converting the maximum value of the sum of the disaster occurrence rates of each of the above characteristic factors to 100. By setting such disaster prediction items, construction work can be performed based on the structure and shape of buildings that differ in the difficulty of construction.
Personnel and equipment that require advance planning, construction methods, usage status of temporary materials to ensure a safe construction environment and equipment to prevent disasters, understanding of work progress by site managers, necessary construction policy understanding and work guidance, Leadership and other evaluations that are factors in the occurrence of disasters are taken into consideration.

Disaster prediction measurement items include 8 items of disaster risk calculated using fixed values and variable values shown in Figures 2 and 3, and numerical values calculated based on past experience based on these evaluation values. Figure 4 shows the characteristics of each workplace group. In Figure 4, for example, the construction difficulty level is 60% to 100.
%, presence or absence of disaster history, and other items are evaluated as 0.
When the average value is shown on a scale of % to 100% for the workplace where the disaster occurred and the workplace where the disaster occurred, the workplace where the disaster occurred is indicated by the dotted line, and the workplace where the disaster occurred is indicated by the solid line. Where are the evaluation values X8 and x of the disaster-occurring work site and operating work site stored in the evaluation value storage units 2 and 3 shown in Figure 1 located in the graph shown in Figure 4 for each of these items? It is a value evaluated based on In other words, taking construction difficulty as an example, whether the construction difficulty of the work site that was the subject of evaluation is closer to the work site where the disaster occurred as shown in Figure 4, that is, closer to the center than between the solid line and the dotted line, or The evaluation value differs depending on whether it is closer to the work place or closer to the outside than the middle between the solid line and the dotted line. Functional calculations are then performed by multiplying such evaluation values by constants A and a.

When performing functional calculations divided into eight items as shown in FIG. 4, evaluation values and constant values are determined, for example, as follows. In other words, the respective evaluation values X1 and xl of the workplace
is the construction difficulty level, X2 and x2 are the presence or absence of disaster history, and X3 and x
3 is the presence or absence of a dangerous construction process, X, and x4 are the implementation status of advance planning, X, and X are the implementation status of the safe construction cycle,
and x6 are the evaluation of the safety management situation, X7 and x7 are the history evaluation of the correction situation, and X8 and x8 are the degree of disaster risk. Then, a function Z+=Ao+ΣA is calculated using the evaluation values of the workplace where the disaster occurred as the corresponding constants A0 to A and a0 to a8,
X, Z,'=a0+Σai X.

and the evaluation value of accident-free workplaces
A value is set such that a bias occurs in the function Z+'=Ao+ΣAHXi'zz'=a0+ΣaiXi'. In other words, when calculating the functions Zl and 22 for workplaces in operation that are in the same situation as the workplace where the disaster occurred, the function 2
1 is more similar to function Z than function 22, zz
The similarity to function 2 is significantly lower than the similarity to function I, but on the other hand, the similarity to function 2 for workplaces that are in the same situation as non-disaster workplaces among working workplaces is lower than that for function z2.
Constants A0 to A8 and a0 to a8 are set to values that are significantly lower than the similarity to t'.

The inventors of the present application set the following value A0 = 84.84378 ao = 69 using the constants A0 to AIl and a0 to a as an example.
．． 04133A+ = 7.88044 a1
= 6.92438A, = 9.96012
2z = 7.35430A3 = 6.296
64 23 = 6.94206A4 = 96
．． 20555 a, = 90.10310A
s = 4.20560 as = 4.518
94A6=4.23298 a6=3.
07496At = 15.77218av
= 14.13361As = -2,88123
2s = -4,33489, and set a reference point between the solid line and the broken line shown in Figure 4 for the evaluation value of each disaster prediction measurement item, and set the outside as "1" and the inside as "2". I tried to evaluate and perform judgment processing. FIG. 5 shows an example of the result of the workplace discrimination determination. Figure 5 indicates disasters, 3 indicates flying/falling, 4 indicates collapse/collapse, 5 indicates electric shock, 6 indicates fire, 7 indicates other disasters, and 0 for each item indicates good, Δ indicates partially defective, and × indicates Indicates general failure. According to this result, the accident occurrence rate from workplaces determined to be accident-free is 12.5%, and the accident occurrence rate from workplaces determined to be accident-prone is 87.5%, achieving extremely high predictive detection ability. I was able to do that. Note that the probability of group determination in FIG. 4 can be determined from the similarity relationship between the distribution of each discriminant function and the group function of each operating workplace.

As explained above, the present invention weights evaluation values of disaster prediction measurement items using constants corresponding to workplaces where disasters have occurred and workplaces where there have been no accidents, and calculates a function calculated using this weighting. It is possible to improve prediction accuracy by determining the group to which each operating workplace belongs based on the group and then determining its probability. It goes without saying that these are not limited to the examples described above, and the number of classification items, the number of evaluation value stages, constant values, etc. may be changed depending on the construction, work place, and other conditions.

〔Effect of the invention〕

As is clear from the above description, according to the present invention, by setting disaster prediction measurement items, weighting the evaluation values, and performing a function process, workplaces can be divided into disaster-occurring workplaces and accident-free workplaces. Since it is determined which group the disaster belongs to, it is possible to search for various factors behind the mechanism of disaster occurrence, and it is possible to improve prediction accuracy. Furthermore, by improving prediction accuracy, it is possible to provide high-quality safety guidance and reduce the number of disasters that occur.

[Brief explanation of drawings]

Fig. 1 is a diagram showing the configuration of one embodiment of the disaster prediction information processing system of the present invention, Fig. 2 is a diagram for explaining an example of fixed values of disaster prediction measurement items, and Fig. 3 is a diagram showing an example of fixed values of disaster prediction measurement items. ! I
Ji IU°**(i(7)ml'fr:mM"7"o
(g゛,; Figure 4 is a diagram for explaining the characteristics of disaster prediction measurement items by workplace group, and Figure 5 is a diagram showing an example of the judgment results output by the disaster prediction information processing system of the present invention. ■...Coefficient storage section, 2 and 3...Evaluation value storage section, 4.
...Discriminant function calculation section, 5...Operating work place function calculation section,
6... Disaster prediction discrimination judgment department, 7... Human resources department.

Claims (3)

[Claims] (1) Disaster prediction measurement items are set, evaluation values of the disaster prediction measurement items are determined for each workplace, and disaster prediction is performed for each operating workplace based on a function obtained by calculating the evaluation value and a predetermined constant. A disaster prediction information processing system that provides information, comprising a discriminant function calculation means for calculating a disaster occurrence workplace group discriminant function and a disaster-free workplace group discriminant function from evaluation values of disaster occurrence workplaces and disaster-free workplaces; and a group function calculation means for calculating a group function for each operating workplace from the evaluation value of the operating workplace, and for determining the similarity between the group function and the accident-occurring workplace group discriminant function and the accident-free workplace group discriminant function. A disaster prediction information processing system characterized by providing disaster prediction information. (2) The disaster prediction information processing system according to claim (1), wherein the discriminant function calculating means calculates the accident-free workplace group discriminant function based on the evaluation values of the operating workplaces. . (3) The group function calculation means calculates a group function for each operating workplace using the same calculation as the accident-occurring workplace group discriminant function and the accident-free workplace group discriminant function, and each group function distinguishes the accident-occurring workplace group. The disaster prediction information processing system according to claim 1, characterized in that the group to which the two objects belong is determined by checking whether they are similar by a function or a disaster-free workplace group discriminant function.