Hazard prediction system

Abstract

To analyze data regarding accumulated industrial accidents and feed back the data for subsequent countermeasures.SOLUTION: A danger prediction system includes an industrial accident information integration server having an industrial accident information database for storing data regarding industrial accidents, and an extraction analysis information database including a classification table for listing classification of factors causing industrial accidents, a factor table for listing factors of industrial accidents extracted from data regarding industrial accidents, and an advice table for correlating pieces of danger prediction advice to check items corresponding to factors, and a check list including check items for checking a worker's business contents, and displays a diagnosis result of risk for a check result, a danger prediction advice related to business contents extracted from the industrial accident information integration server, and data regarding industrial accidents on a terminal operated by the worker when the check result checked by the worker is transmitted to the industrial accident information integration server.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present disclosure relates to a danger prediction system.

Background Art

[0002] There is a system that searches for and provides past disasters, accidents, etc. For example, there is a technology related to an accident prevention management method for determining necessary and sufficient accident recurrence prevention measures at important management points in appropriate work processes (see Patent Document 1). In this technology, the determination of accident recurrence prevention measures is configured to prioritize the accident recurrence prevention measures for operations that may cause highly dangerous accidents according to the risk level.

[0003] Also, there is a technology related to a danger prediction support system that provides appropriate information for preventing the occurrence of industrial accidents (see Patent Document 2). In this technology, a value obtained by multiplying the risk level of the extracted danger point by the near-miss index is calculated as the actual risk level, and a model danger prediction table describing the danger points with a high actual risk level and their countermeasures is created.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] In such a technical field, there is a need for a system that analyzes data related to industrial accidents and provides feedback for countermeasures. For example, at the work site of a chemical factory, accident prevention is attempted in daily efforts such as KY (hazard prediction) activities. In such efforts, it is conceivable to provide an opportunity to learn from past industrial accident cases at other sites and provide feedback. It is assumed to provide an opportunity to recognize risks that are not usually noticed, including past industrial accident cases at other sites and human factors.

[0006] It is also assumed that the daily self-checks of workers are analyzed so that managers can check the analysis results, grasp the ideas and problems at the site, and provide feedback for countermeasures.

[0007] An object of the present disclosure is to provide a hazard prediction system that can analyze accumulated data related to industrial accidents and provide feedback for subsequent countermeasures.

Means for Solving the Problems

[0008] The hazard prediction system of the present disclosure includes a labor accident information database in which data related to industrial accidents (labor accident information) is stored, a classification table describing the classification of factors causing labor accidents, a factor table describing the factors of labor accidents extracted from the data related to industrial accidents, and an advice table in which hazard prediction advice is associated with check items corresponding to the factors, and a labor accident information aggregation server including an extraction analysis information database, and a checklist including the check items for checking the work content of workers. When the worker scores and checks the relevance between the content of the check items described in the checklist and his / her own work content and transmits the check result to the labor accident information aggregation server, the diagnosis result of the risk for the check result, the hazard prediction advice related to the work content extracted from the labor accident information aggregation server, and the data related to the industrial accident are displayed on the terminal operated by the worker.

[0009] Further, in the risk prediction system of the present disclosure, the aggregated result of the check results aggregated for each check may be output to a terminal operated by an administrator.

[0010] Further, in the risk prediction system of the present disclosure, the classification table has a combination of a classification that classifies the awareness of workers at the time of industrial accidents and actions / situations leading to industrial accidents, and in the extraction analysis information database, the factors in the factor table may be recorded for the combination.

[0011] Further, in the risk prediction system of the present disclosure, it may be configured to receive the type of work planned by the worker and output the check list including the check items extracted from a list of check items determined in advance according to the type of work.

[0012] Further, in the risk prediction system of the present disclosure, the check list may be configured to display different check items for each work unit for the work continuously performed at the work site.

Advantages of the Invention

[0013] According to the risk prediction system of the present disclosure, an effect can be obtained that accumulated data on industrial accidents can be analyzed and fed back to subsequent countermeasures.

Brief Description of the Drawings

[0014]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Mode for Carrying Out the Invention

[0015] Hereinafter, embodiments of the technology of the present disclosure will be described with reference to the drawings.

[0016] First, the outline of the danger prediction system proposed in the present disclosure will be described. FIG. 1 is a diagram schematically showing the main steps of the danger prediction system. As shown in FIG. 1, the danger prediction system is a mechanism for supporting the activities of each of the factory managers and workers regarding danger prediction. In step 1, the manager inputs data related to past industrial accidents into the danger prediction system, and stores the data related to industrial accidents (hereinafter also referred to as industrial accident information) in the industrial accident information database (DB) of the danger prediction system. In step 2, the manager extracts the actions leading to industrial accidents and the factors causing the actions, and records the analysis results of the data related to industrial accidents in the danger prediction system. In step 3, the manager inputs KY advice in comparison with the analysis results of the recorded data related to industrial accidents, and records it in various tables of the danger prediction system. A self-check list for the worker to check the work content is generated. In step 4, as part of the daily KY activities, the worker performs checks using the output self-check. The worker checks by scoring the relevance between the content of the check items described in the check list and his / her own work content. In step 5, based on the check results received by the danger prediction system, the diagnosis result of the risk for the check results, KY advice and related industrial accident information related to the work content are extracted and displayed on a terminal that the worker can confirm. In step 6, the daily check results of the worker are aggregated for use by the manager in industrial accident risk management.

[0017] Next, the functional configuration of the danger prediction system 100 will be described. FIG. 2 is a block diagram showing an example of the functional configuration of the danger prediction system 100 of the present embodiment. As shown in FIG. 2, in the danger prediction system 100, an industrial accident information aggregation server 110, a manager terminal 140, and a worker terminal 150 are connected via a network N. The industrial accident information aggregation server 110 includes an industrial accident information DB 112, an extraction analysis information DB 114, a diagnosis processing unit 116, and a tabulation processing unit 118. Note that it is assumed that a plurality of terminals are installed for each production site for the manager terminal 140 and the worker terminal 150. Also, when there is a site that oversees each production site, a manager terminal 140 is also installed at that site.

[0018] The danger prediction system 100 is assumed to be used, for example, by aggregating and storing data related to industrial accidents at each production site of a factory. As a result, past industrial accident cases for each production site are aggregated and can be used for further analysis. Also, by storing the self-check list and KY advice described below, the lessons learned from past industrial accidents can be conveyed. Also, it becomes possible to take corresponding measures according to each production site.

[0019] The manager terminal 140 and the worker terminal 150 are terminals for supporting danger prediction in cooperation with the industrial accident information aggregation server 110. The manager terminal 140 is a PC terminal or the like that can be operated by a manager, and is a terminal for inputting and outputting data to and from the industrial accident information aggregation server 110. As input and output of data to and from the industrial accident information aggregation server 110, the manager terminal 140 can input industrial accident information and KY advice, etc. Also, the manager terminal 140 can display the tabulation of the check results recorded in the industrial accident information aggregation server 110. The worker terminal 150 is a PC terminal or the like that can be operated by a worker. The worker terminal 150 can display the self-check list on a display or the like, and transmits the check results of the self-check list checked by the worker to the industrial accident information aggregation server 110.

[0020] In the following description, the content implemented mainly by the administrator or the operator is implemented using the administrator terminal 140 or the operator terminal 150. Note that the operator may include a work supervisor.

[0021] The configuration of the industrial accident information aggregation server 110 as hardware is shown in FIG. 3. It has a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a storage 14, an input unit 15, a display unit 16, and a communication interface (I / F) 17. Each configuration is communicably connected to each other via a bus 19.

[0022] The CPU 11 is a central processing unit that executes various programs and controls each part. That is, the CPU 11 reads a program from the ROM 12 or the storage 14 and executes the program using the RAM 13 as a work area. The CPU 11 performs control of each of the above configurations and various arithmetic processes according to the programs stored in the ROM 12 or the storage 14. In the present embodiment, programs for analysis processing are stored in the ROM 12 or the storage 14.

[0023] The ROM 12 stores various programs and various data. The RAM 13 temporarily stores programs or data as a work area. The storage 14 is composed of a storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive), and stores various programs including an operating system and various data.

[0024] The input unit 15 includes a pointing device such as a mouse, and a keyboard or a microphone for voice input, and is used to perform various inputs.

[0025] The display unit 16 is, for example, a liquid crystal display, and displays various information. The display unit 16 may adopt a touch panel method and function as the input unit 15.

[0026] The communication interface 17 is an interface for communicating with other devices such as terminals. For such communication, for example, standards for wired communication such as Ethernet (registered trademark) or FDDI, or standards for wireless communication such as 4G, 5G, or Wi-Fi (registered trademark) are used.

[0027] The functions and each component of the industrial accident information aggregation server 110 will be described.

[0028] The industrial accident information DB 112 is a database in which data (industrial accident information) regarding industrial accidents received from the administrator terminal 140 is stored. The data regarding industrial accidents is information that records the details of each occurrence case of an industrial accident. An example of the items of the table recorded as data regarding industrial accidents will be given. As the items of the table, for example, specific work, factory, date of occurrence, type of industrial accident, outline of industrial accident, place of occurrence, name of injury or illness, type of accident, work number, work category, victim information, causative agent, disaster occurrence situation, situation image, direct cause, elemental cause, and main countermeasures, etc. are used as an example. For each occurrence case of an industrial accident, the contents of these items are recorded in the data regarding industrial accidents. From the administrator terminal 140 installed for each production base, data regarding industrial accidents for each production base, that is, for each factory, is input and accumulated.

[0029] An example of the "disaster occurrence situation" for the data regarding industrial accidents will be given. For example, in the "disaster occurrence situation", there is recorded an explanation of a situation such as "(**month **day), night shift, while the victim was alone making a film with **machine in **factory, the holding roll in the state of the image was floating, so when trying to press it with bare hands, the holding roll rose in the direction of the arrow, and as a result, the right hand that was pressing the roll was pinched between the angle."

[0030] The extraction and analysis information DB 114 is a database that includes analysis data obtained by analyzing data regarding industrial accidents and master data of a self-check list generated based on the analysis.

[0031] An example of the analysis data is shown in Fig. 4. As tables of the analysis data (AD), there are a classification table (A1), a count table (A2) showing the number of occurrences and frequencies, and a factor table (A3) showing factors (related words). In terms of format, for each combination of classification (A1-1) and action / situation (A1-2), information is recorded in the count table (A2) showing the number of occurrences and frequencies and the factor table (A3) showing factors (related words).

[0032] Note that among the tables, the information in the classification table (A1) has been set in advance by the administrator with classifications assumed for industrial accidents. Among the tables, the information in the count table (A2) and the factor table (A3) is the information obtained by the administrator extracting and analyzing data related to industrial accidents from the industrial accident information DB112. The extraction / analysis is performed, for example, by extracting the content such as the "accident occurrence situation" of the data related to industrial accidents and grouping synonymous words. However, the analysis data is not limited to the case where the administrator performs extraction / analysis and records it. Based on a model learned by machine learning or the like, extraction / analysis may be automatically performed, and the administrator may confirm the result and record it as the analysis data. The learning may be to train the model to extract factors using the data related to industrial accidents as input.

[0033] The classification table (A1) is a table describing the classification of the factors causing industrial accidents in the data related to industrial accidents. The classification table (A1) is divided into a classification (A1-1) and an action / situation (A1-2) leading to industrial accidents as sub-tables. The classification (A1-1) classifies the awareness of the operator (human factor) at the time of the industrial accident and is represented by any of the seven classifications (careless, ignoring, reckless, unconscious, ignorant, unaware, insensitive, unconcerned, unauthorized). Note that the classification is just an example, and it may be appropriately added, deleted, or changed according to the factory.

[0034] Action / Situation (A1-2) is a sub-table that describes actions / situations leading to industrial accidents. It can be said that actions / situations summarize the specific factors in the Factor Table (A4). As shown in Figure 4, "not performing necessary operations / confirmations", "not working according to the specified procedures", "putting hands on rotating bodies / operating equipment", etc. are described as actions / situations. Note that actions / situations are just examples and may be appropriately added, deleted, or modified according to the factory.

[0035] In addition, the classification in the Classification Table (A1) and the content of actions / situations may be reset based on the content of the updated industrial accident information DB112. For example, some classifications or some actions / situations may be grouped together.

[0036] The Case Number Table (A2) is a table that describes the number of occurrences for each case of the factors extracted from the industrial accident information DB112 and the frequency (%) with respect to the total number of the extracted occurrence cases.

[0037] The Factor Table (A3) is a table that lists the factors of industrial accidents extracted from the industrial accident information DB112. As shown in Figure 4, the "Factor (Related Words)" which is the Factor Table (A3) lists contents such as "familiar work (6), urgent request, assembling the rolling tower immediately before lunch break and being in a hurry, being worried that the work would be delayed, and the risk awareness decreasing due to familiar work...".

[0038] Figure 5 shows an example of the master data of the self-check list. The master data of the self-check list is divided into the self-check list itself (A) and the KY advice (B) for the check items. The table of KY advice (B) is an example of the advice table of the technology of the present disclosure.

[0039] Regarding the check items (A) of the self-check list, the classification and actions / situations are the same as the analysis data. For the check items, the administrator sets them based on the content of the factor table (A3) of the analysis data, etc. The KY advice (B) is set by the administrator entering the KY advice according to the check items set for the self-check list.

[0040] As shown in FIG. 5, the KY advice shows descriptions such as "Review and confirm the operations even for familiar work" for the check item "Today's work is familiar work" and "Make sure to confirm the work procedures especially in case of emergency" for the check item "Currently receiving an urgent request".

[0041] Note that the check items of the self-check list may be automatically generated based on a model learned by machine learning or the like, and the administrator may confirm and set the results. For learning, the analysis data and the check items of the master data are used as teacher data, and the model may be learned to input the analysis data and output the check items. Also, a model for classifying which check items the KY advice corresponds to may be learned, and the model may be used to associate the check items with the KY advice. Further, the model may be used to present candidates for the association to the administrator and let the administrator select an appropriate association.

[0042] Also, the check items of the self-check list are not limited to the same classification and actions / situations as the analysis data, and may be classified by work type of the data related to industrial accidents in the industrial accident information DB112. Thereby, the danger prediction system 100 can receive the work type that the worker plans to perform and output a self-check list including the check items extracted from the check item list determined in advance according to the work type to the worker terminal 150. The check item list is a list of candidates for check items for each work type.

[0043] The diagnostic processing unit 116 creates a self-check list from the master data in the extraction analysis information DB 114 and outputs it to the operator terminal 150. The self-check list only needs to be configured so that the operator can score and check the relevance between the content of the check items described in the self-check list and their own work content. Further, the administrator may be able to set the self-check list so that it includes check items that are emphasized in the work assumed in the factory work.

[0044] An example of the self-check list is shown in FIG. 6. As shown in FIG. 6, the self-check list 180 has a selection column for processes and operations and is selectable. The self-check list 180 is a list for checking the degree represented by "different" or "as it is" for the check items with numbers from 1 to 5. The operator checks the numbers of the check items in the self-check list 180 according to the work content of the planned work.

[0045] In addition, the self-check list may be configured to display different check items for each work unit, that is, every time a work check occurs, for the work continuously performed at the work site. Also, different check items may be displayed according to the selected process and operation. This can prevent the operator from checking randomly and promote accident prevention.

[0046] When the diagnosis processing unit 116 receives the check result of the self-check list from the operator terminal 150, it generates a risk diagnosis result for the check result. The generation of the diagnosis result may be performed using a rule or a learned model in which the numerical values of the check results are associated with the diagnosis results in advance. Further, the diagnosis processing unit 116 extracts related KY advice from the master data of the extraction analysis information DB 114 and extracts related industrial accident information from the industrial accident information DB 112 based on the work content included in the check result. The diagnosis processing unit 116 outputs the generated risk diagnosis result, the extracted KY advice, and the related industrial accident information to the operator terminal 150 for display. As the object to be extracted as the related industrial accident information, for example, data related to labor accidents at other production bases may be extracted.

[0047] FIG. 7 shows an example of the diagnosis result of the self-check list and the like. As shown in FIG. 7, the diagnosis result is displayed in a chart showing numerical values for each classification, and the description of the KY advice and the list of related industrial accident information are displayed. By displaying the diagnosis result in this way, an opportunity can be provided to notice risks that are not usually noticed in the self-check. In addition, effective KY activities can be carried out based on the KY advice. Further, by displaying the related industrial accident information, an opportunity can be provided to learn about past industrial accident cases including those at other bases.

[0048] In addition, the overall trend may be recorded from the aggregation of the check results, and the comparison between the operator's trend and the overall trend may be shown in the diagnosis result chart.

[0049] In addition, the diagnosis function may be configured as software in which modules are combined so that it can be executed on the operator terminal 150.

[0050] The aggregation processing unit 118 aggregates the check results collected for each check by the operator and outputs the aggregation result to the administrator terminal 140. Note that the work unit is each check for which the operator has checked the self-check list. In this way, by enabling the administrator to confirm the aggregation result of the self-check, it becomes possible to grasp what kind of awareness the operator has. Also, the scope of aggregation may be by department. This makes it possible to grasp the risks and issues for each department.

[0051] Fig. 8 shows an example of the aggregation result of the check results. In the aggregation in Fig. 8, the check results are aggregated for each record in the unit of the check date, and numerical checks for the check items are displayed, which is an example of displaying aggregation results such as the total and average.

[0052] Next, with reference to Fig. 9, the operation of the industrial accident information aggregation server 110 in the industrial accident prediction system 100 will be described. By the CPU 11 reading a program from the ROM 12 or the storage 14 and expanding and executing it in the RAM 13, the CPU 11 functions as each part of the industrial accident information aggregation server 110, and the following processing is performed.

[0053] In step S100, data related to industrial accidents (industrial accident information) is received from the administrator terminal 140, and the industrial accident information is stored in the industrial accident information DB 112.

[0054] In step S102, extraction / analysis of data related to industrial accidents (industrial accident information) is performed by various operations of the administrator on the administrator terminal 140, and the analysis data is recorded in the extraction analysis information DB 114.

[0055] In step S104, a self-check list is generated by various operations of the administrator on the administrator terminal 140, and the master data of the self-check list is recorded in the extraction analysis information DB 114.

[0056] In step S106, a self-check list is created from the master data in the extraction analysis information DB114, and the self-check list is output to the worker terminal 150.

[0057] In step S108, the check result of the worker's self-check list is received from the worker terminal 150.

[0058] In step S110, a risk diagnosis result for the check result is generated, relevant KY advice is extracted from the master data of the self-check list in the extraction analysis information DB114, and relevant industrial accident information is extracted from the industrial accident information DB112.

[0059] In step S112, the generated risk diagnosis result, the extracted KY advice and relevant industrial accident information are output to the worker terminal 150 and displayed on the worker terminal 150.

[0060] In step S114, the check results collected for each check by the worker are totaled, and the total result is output to the administrator terminal 140.

[0061] As described above, according to the risk prediction system 100 according to the present embodiment, the accumulated data on industrial accidents can be analyzed and fed back to subsequent countermeasures.

[0062] Note that the present invention is not limited to the above-described embodiments, and various modifications and applications are possible without departing from the gist of the present invention.

[0063] Also, in the present specification, although the embodiment in which the program is pre-installed has been described, it is also possible to store the program in a computer-readable recording medium and provide it.

Explanation of Signs

[0064] 100 Risk prediction system 110 Industrial accident information aggregation server 112 Industrial Accident Information DB 114 Extraction and Analysis Information DB 116 Diagnosis Processing Unit 118 Aggregation Processing Unit 140 Administrator Terminal 150 Operator Terminal 180 Self-Check List

Claims

1. An industrial accident information database storing data related to industrial accidents, an extraction and analysis information database including a classification table describing the classification of factors causing industrial accidents, a factor table describing the factors of industrial accidents extracted from the data related to industrial accidents, and an advice table associating risk prediction advice with check items corresponding to the factors, and an industrial accident information aggregation server including the same, a checklist including the check items for checking the work content of workers, when the worker sends a check result obtained by scoring and checking the relevance between the content of the check items described in the checklist and his / her own work content to the industrial accident information aggregation server, a diagnosis result of the risk for the check result, the risk prediction advice related to the work content extracted from the industrial accident information aggregation server, and the data related to the industrial accident are displayed on a terminal operated by the worker. A risk prediction system.

2. The risk prediction system according to claim 1, wherein an aggregation result of the check results aggregated for each check is output to a terminal operated by an administrator.

3. The classification table has a combination of a classification classifying the consciousness of workers at the time of industrial accident occurrence and an action / situation leading to an industrial accident, The risk prediction system according to claim 1 or claim 2, wherein in the extraction and analysis information database, the factors in the factor table are recorded for the combination.

4. The risk prediction system according to any one of claims 1 to 3, which receives a work type planned to be performed by the worker and outputs the checklist including the check items extracted from a checklist of check items determined in advance according to the work type.

5. The risk prediction system according to any one of claims 1 to 4, wherein the checklist displays different check items for each work unit for the work continuously performed at the work site.

Images