Work risk evaluation recommendation device, method, computer-readable recording medium, and computer program

Abstract

This work risk evaluation recommendation method comprises the steps of: receiving an input of a worker text which is at least one text corresponding to work to be performed by a worker; standardizing the worker text into a standard text which is a text for work risk evaluation; generating a candidate list including at least one work risk evaluation format corresponding to the standard text; and generating a recommendation list which is at least one work risk evaluation format, to be recommended to a user, in the candidate list on the basis of a similarity analysis of the standard text.

Description

Recommended work risk assessment device, method, computer-readable recording medium and computer program

[0001] The present invention relates to a technique for recommending a work risk assessment that matches the work to be performed by a worker.

[0002] In order to identify risk factors at each work stage and eliminate potential risks at work sites such as factories, a job risk assessment (JSA) is created. However, these job risk assessments are burdensome for workers because the job risk assessments that must be considered vary depending on the work site or work content.

[0003] (Prior art literature)

[0004] (Patent Document 1) Republic of Korea Patent Publication No. 10-2012-0087784: Safety management system and method for mandatory work safety analysis

[0005] An object of the present invention is to provide a technique for recommending a work risk assessment that is suitable for the work to be performed by a worker.

[0006] A method performed by a work risk assessment recommendation device according to one aspect of the present invention comprises the steps of: receiving a worker text, which is at least one text corresponding to a work to be performed by a worker; standardizing the worker text into a standard text, which is a text for work risk assessment; generating a candidate list including at least one work risk assessment format corresponding to the standard text; and generating a recommendation list, which is at least one work risk assessment format to be recommended to a user, from among the candidate list based on a similarity analysis of the standard text.

[0007] Additionally, the worker text may include text corresponding to at least one of a work location, worker affiliation, work target, tool utilized, and work type.

[0008] Additionally, the step of standardizing to a standard text can be standardized by retrieving at least one text corresponding to each text included in the worker text from a database containing a plurality of standard texts.

[0009] Additionally, the step of normalizing to a standard text may generate a registration request for text that does not exist in the database if a text corresponding to the text included in the worker text does not exist in the database.

[0010] Additionally, the step of standardizing to a standard text may further include extracting a superordinate concept text of the standard text based on the standard text.

[0011] Additionally, the step of generating a candidate list may generate the candidate list by searching a database including a plurality of job risk assessments for at least one job risk assessment format corresponding to the upper concept text.

[0012] Additionally, the standard text may include at least one of the following: work location, self-employed affiliation, work target, tools utilized, and type of work.

[0013] Additionally, the superordinate concept text may include at least one of the superordinate concept of the work object, the superordinate concept of the tool used, and the superordinate concept of the type of work.

[0014] Additionally, the step of generating a candidate list may generate a candidate list by converting at least one task risk assessment format corresponding to the upper concept text into a word form or sentence form of the upper concept text.

[0015] According to another aspect of the present invention, a work risk assessment recommendation device comprises a memory including a command; and a processor that performs a method including the steps of: receiving a worker text, which is at least one text corresponding to a task to be performed by a worker, by executing the command; standardizing the worker text into a standard text, which is a text for work risk assessment; generating a candidate list including at least one work risk assessment format corresponding to the standard text; and generating a recommendation list, which is at least one work risk assessment format to be recommended to a user, from among the candidate list based on a similarity analysis of the standard text.

[0016] According to one aspect of the present invention, it becomes possible to recommend a work risk assessment that matches the work to be performed by a worker.

[0017] FIG. 1 is a block diagram of a work risk assessment recommendation device according to one embodiment of the present invention.

[0018] Figure 2 is a flowchart of a work risk assessment recommendation method according to one embodiment of the present invention.

[0019] Figures 3 to 13 are diagrams showing examples for explaining a work risk assessment recommendation process according to one embodiment of the present invention.

[0020] Figure 14 is a block diagram of a work risk assessment recommendation device according to another embodiment of the present invention.

[0021] The advantages and features of the present invention, and the methods for achieving them, will become clearer with reference to the embodiments described in detail below with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various forms. These embodiments are provided solely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention, and the scope of the present invention is defined solely by the claims.

[0022] In describing embodiments of the present invention, specific descriptions of known functions or configurations will be omitted unless actually necessary. Furthermore, the terms described below are defined based on their functions in the embodiments of the present invention and may vary depending on the intent or custom of the user or operator. Therefore, their definitions should be based on the overall content of this specification.

[0023] The terms ‘…bu’, ‘…gi’, etc. used hereinafter mean a unit that processes at least one function or operation, which can be implemented by hardware, software, or a combination of hardware and software.

[0024] FIG. 1 is a block diagram of a work risk assessment recommendation device according to one embodiment of the present invention.

[0025] Referring to FIG. 1, a work risk assessment recommendation device (1000) according to one embodiment of the present invention may include an input unit (1100), a text standardization unit (1200), a candidate list generation unit (1300), a recommendation list generation unit (1400), a database (1500), and an output unit (1600).

[0026] The input unit (1100) can receive data used for work risk assessment recommendations from an external device (not shown) using wired or wireless communication.

[0027] In one embodiment, the input unit (1100) can receive a plurality of work risk assessment formats, and the plurality of work risk assessment formats can be different depending on the type of company, work location, worker affiliation, work target, tool used, work type, etc.

[0028] In one embodiment, the input unit (1100) may receive standard text related to a task risk assessment. Specifically, the standard text may include standard text (e.g., "Reactor Channel") for a worker text (e.g., "Reactor Top Channel") that the worker inputs to search for a task risk assessment format corresponding to the task to be performed.

[0029] Additionally, the input unit (1100) can receive data required for recommending a work risk assessment from a user using a user interface (e.g., keyboard, touch screen), etc.

[0030] In one embodiment, the input unit (1100) may receive text, i.e., worker text, from a worker to search for a work risk assessment format corresponding to a task to be performed by the worker.

[0031] In one embodiment, worker text may include the type of business, work location, worker affiliation, work target, tool utilized, type of work, etc.

[0032] The text standardization unit (1200) can standardize worker text input from workers into standard text for evaluating work stability. Here, the standard text can include Korean, English, etc.

[0033] In one embodiment, the text standardization unit (1200) can standardize the worker text into a standard text by searching the database (1500) for a standard text corresponding to the worker text.

[0034] In one embodiment, the text standardization unit (1200) may output a request for registration of a new standard text if a standard text corresponding to the worker text is not stored in the database (1500). The work risk assessment recommendation device (1000) may output a request for registration of a standard text on a display unit (not shown) or transmit the request to a server (not shown) that manages the work risk assessment recommendation device (1000), thereby allowing the administrator of the work risk assessment recommendation device (1000) to recognize that registration of a standard text is necessary. Thereafter, the administrator may perform an update task for adding a new standard text to the database (1500).

[0035] In one embodiment, the text standardization unit (1200) can convert worker text into standard text using a natural language processing model such as GPT.

[0036] In one embodiment, the text standardization unit (1200) can convert each worker text into a standard text by considering the characteristics of each worker text. Specifically, if the worker text relates to a location, the text standardization unit (1200) can search for a standard text corresponding to the worker text among the standard texts related to the location.

[0037] When the worker text is converted into standard text, the candidate list generation unit (1300) can generate a candidate list including multiple work risk assessment formats corresponding to the standard text based on the standard text.

[0038] In one embodiment, the candidate list generation unit (1300) can generate a list by searching for at least one work risk assessment format among a plurality of work risk assessment formats stored in the database (1500) based on the word form of the standard text, the sentence form composed of the standard text, etc.

[0039] In one embodiment, the candidate list generation unit (1300) can search for at least one work risk assessment format including text corresponding to at least one of a standard text and a superordinate concept text in the database (1500), and generate a candidate list including the searched at least one work risk assessment format.

[0040] In one embodiment, the candidate list generation unit (1300) may output text included in at least one work risk assessment format that includes text corresponding to at least one of standard text and upper concept text, and generate a candidate list in the form of words of the extracted text or sentences composed of the extracted text.

[0041] In one embodiment, the candidate list generation unit (1300) may generate a recommendation list using information such as team information, work subject, work target, work tool, and work type. Team information and work subject may refer to information used for the purpose of hard filtering / sorting specific to the field domain (e.g., in the case of own work, using information only among own works, etc.). Work target, work tool, work type, etc. are used to calculate each cosine distance before being collated into each word embedding form, and then a list can be generated by calculating a weighted average of the three similarity scores (Ex: Weighted average distance = work target similarity * 0.5 + work type similarity * 0.4 + work tool similarity * 0.1).

[0042] The recommendation list generation unit (1400) can generate a recommendation list including at least one work risk assessment format to be recommended to a worker among at least one work risk assessment format included in the recommendation list based on standard text.

[0043] In one embodiment, the recommendation list generation unit (1400) can generate a recommendation list from a candidate list based on word similarity, sentence similarity, etc. of standard text.

[0044] In one embodiment, the recommendation list generation unit (1400) can use a language processing model such as LLM to calculate word similarity or similarity based on a combination of words.

[0045] In one embodiment, when a command to select at least one of at least one work risk assessment format output from a worker is input from the worker, the recommendation list generation unit (1400) can finally determine the work risk assessment format corresponding to the input command as the worker's work risk assessment format.

[0046] In one embodiment, the recommendation list generation unit (1400) may generate a recommendation list by utilizing sentence embedding information for the entire standardized sentence itself in addition to the task target and task type information. At this time, the recommendation list may be generated based on a weighted average through a weighted combination of sentence similarity and word similarity (Ex. Weighted average distance = sentence similarity * 0.3 + task type similarity * 0.25 + task target similarity * 0.45).

[0047] The database (1500) may include data used for recommending work risk assessments.

[0048] In one embodiment, the database (1500) may include multiple work risk assessment formats, and the multiple work risk assessment formats may differ depending on the type of business, work location, worker affiliation, work target, tool utilized, work type, etc.

[0049] In one embodiment, the database (1500) may include standard text related to job risk assessments. Specifically, the standard text may include standard text (e.g., "Reactor Channel") for worker text (e.g., "Reactor Top Channel") that a worker inputs to search for a job risk assessment format corresponding to the task to be performed.

[0050] The output unit (1600) can output a recommendation list so that the worker can recognize it. Specifically, the output unit (1600) can output at least one work risk assessment format included in the recommendation list to a display unit (not shown).

[0051] Figure 2 is a flowchart of a work risk assessment recommendation method according to one embodiment of the present invention.

[0052] Hereinafter, the above method is described as an example performed by the work risk assessment recommendation device (1000) illustrated in FIG. 1.

[0053] Referring to FIG. 2, in step S2100, the work risk assessment recommendation device (1000) can create a database (1500) including data used to recommend work risk assessment.

[0054] In one embodiment, the database (1500) may include multiple work risk assessment formats, and the multiple work risk assessment formats may differ depending on the type of business, work location, worker affiliation, work target, tool utilized, work type, etc.

[0055] In one embodiment, the database (1500) may include standard text related to job risk assessments. Specifically, the standard text may include standard text (e.g., "Reactor Channel") for worker text (e.g., "Reactor Top Channel") that a worker inputs to search for a job risk assessment format corresponding to the task to be performed.

[0056] In step S2200, the work risk assessment recommendation device (1000) can receive text input from the worker to search for a work risk assessment format corresponding to the work to be performed by the worker, i.e., worker text.

[0057] In one embodiment, worker text may include the type of business, work location, worker affiliation, work target, tool utilized, type of work, etc.

[0058] In step S2300, the work risk assessment recommendation device (1000) can standardize the worker text input from the worker into a standard text for work safety assessment.

[0059] In one embodiment, the work risk assessment recommendation device (1000) can standardize worker text into standard text by searching for standard text corresponding to worker text in a database (1500).

[0060] In step S2400, if a standard text corresponding to the worker text is not stored in the database (1500), the work risk assessment recommendation device (1000) can output a request for registration of a new standard text. The work risk assessment recommendation device (1000) can output the standard text registration request on a display unit (not shown) or transmit it to a server (not shown) that manages the work risk assessment recommendation device (1000), thereby allowing the administrator of the work risk assessment recommendation device (1000) to recognize that registration of the standard text is necessary. Thereafter, the administrator can perform an update task for adding a new standard text to the database (1500).

[0061] In step S2500, when the worker text is converted into standard text, the work risk assessment recommendation device (1000) can generate a candidate list including a plurality of work risk assessment formats corresponding to the standard text based on the standard text.

[0062] In one embodiment, the work risk assessment recommendation device (1000) can generate a list by searching for at least one work risk assessment format among a plurality of work risk assessment formats stored in a database (1500) based on the word form of the standard text, the sentence form composed of the standard text, etc.

[0063] In step S2600, the work risk assessment recommendation device (1000) can generate a recommendation list including at least one work risk assessment format to be recommended to the worker among at least one work risk assessment format included in the recommendation list based on standard text.

[0064] In one embodiment, the work risk assessment recommendation device (1000) can generate a recommendation list from a candidate list based on word similarity, sentence similarity, etc. of standard text.

[0065] In step S2700, the work risk assessment recommendation device (1000) can output a recommendation list so that the worker can recognize it. Specifically, the work risk assessment recommendation device (1000) can output at least one work risk assessment format included in the recommendation list on a display unit (not shown). The work risk assessment recommendation device (1000) can receive a command from the worker to select at least one of the at least one work risk assessment format output. The work risk assessment recommendation device (1000) can finally determine the work risk assessment format corresponding to the command input from the worker as the worker's work risk assessment format.

[0066] Figures 3 to 13 are diagrams showing examples for explaining a work risk assessment recommendation process according to one embodiment of the present invention.

[0067] Referring to Figure 3, an example of a work risk assessment format (or work safety analysis, JSA) recommended by a work risk assessment device according to one embodiment of the present invention is illustrated. It can be seen that the work risk assessment format may vary depending on the work location, company (partner), work title, work content, tools, etc.

[0068] Referring to FIG. 4, the work risk assessment recommendation device (1000) can output a graphical user interface on a display unit (not shown) that allows the worker to input work text, such as a work process, a JSA evaluation form name, and JSA work content. The worker can input content corresponding to the work process, the JSA evaluation form name, and JSA work content, using the user interface, etc.

[0069] In one embodiment, the work process may refer to the names of various production processes in operation at the factory, such as Copolyester, CHDM, SCP, etc.

[0070] In one embodiment, the JSA work evaluation name may refer to a superordinate topic of work to be performed, such as Copoly plate heat exchanger external assembly work (ERC).

[0071] In one embodiment, the JSA work content may refer to the details of the work to be performed, such as the work of removing and installing a plate heat exchanger using impact and hand tools.

[0072] Referring to FIG. 5, an example of converting worker text into standard text is illustrated. If the worker text is CHDM, reactor upper Channel, tensioner hoist, and removal / assembly work, the work risk assessment recommendation device (1000) can search for a standard text corresponding to a synonym of the work text in the database (1500). Reactor Channel corresponding to the reactor upper Channel, Tensioner Hoist corresponding to the tensioner hoist, and Disassembly and assembly corresponding to the removal / assembly work are stored in the database (1500), and if there is no standard text corresponding to CHDM, the work risk assessment recommendation device (1000) can convert reactor upper Channel, tensioner hoist, and removal / assembly work into their respective standard texts, and can notify the worker or manager that a new standard keyword registration is necessary because there is no corresponding word for CHDM.

[0073] Referring to Figure 6, an example of converting a standard text into a superordinate concept text is shown. The work risk assessment recommendation device (1000) can extract a superordinate concept for a standard text by searching the database (1500) for a superordinate concept text corresponding to each standard text.

[0074] Referring to Figure 7, an example of extracting a standard text and its superordinate concepts is shown. The superordinate concepts for the standard text are indicated by adding a group.

[0075] Referring to FIG. 8, an example of a candidate list is shown. The work risk assessment recommendation device (1000) can generate a candidate list including a list of at least one work risk assessment format selected based on a standard text and its higher-level concepts in word or sentence form.

[0076] Referring to Figure 9, a diagram showing an example of converting a work risk assessment format into sentences based on standard text and upper concept text is shown.

[0077] Referring to Figure 10, a diagram showing an example of converting a work risk assessment format into an arrangement of words based on standard text and upper concept text is shown.

[0078] Referring to FIG. 11, an example of calculating cosine similarity with word embeddings of a work risk assessment format stored in a database (1500) based on word embeddings of standard texts by work type is shown.

[0079] Referring to Figure 12, an example of calculating the distance between character embeddings using a recommendation list in sentence form is shown.

[0080] Referring to Figure 13, an example of a recommendation list provided to a worker by a work risk assessment recommendation device is shown.

[0081] Figure 14 is a block diagram of a work risk assessment recommendation device according to another embodiment of the present invention.

[0082] As illustrated in FIG. 14, the work risk assessment recommendation device (1000) may include at least one element among a processor (14100), a memory (14200), a storage unit (14300), a user interface input unit (14400), and a user interface output unit (14500), which may communicate with each other via a bus (14600). In addition, the work risk assessment recommendation device (1000) may also include a network interface (14700) for connecting to a network. The processor (14100) may be a CPU or a semiconductor device that executes processing instructions stored in the memory (14200) and / or the storage unit (14300). The memory (14200) and the storage unit (14300) may include various types of volatile / non-volatile storage media. For example, the memory may include a ROM (14240) and a RAM (14250).

[0083] The devices described above may be implemented as hardware components, software components, and / or a combination of hardware components and software components. For example, the devices and components described in the embodiments may be implemented using one or more general-purpose computers or special-purpose computers, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing instructions and responding. The processing device may execute an operating system (OS) and one or more software applications running on the operating system.

[0084] Additionally, the processing device may access, store, manipulate, process, and generate data in response to the execution of software. For ease of understanding, the processing device is sometimes described as being used alone; however, those skilled in the art will appreciate that the processing device may include multiple processing elements and / or multiple types of processing elements. For example, the processing device may include multiple processors, or a processor and a controller. Additionally, other processing configurations, such as parallel processors, are also possible.

[0085] Software may include a computer program, code, instructions, or a combination of one or more of these, and may configure a processing device to perform a desired operation or, independently or collectively, command the processing device. The software and / or data may be permanently or temporarily embodied in any type of machine, component, physical device, virtual equipment, computer storage medium or device, or transmitted signal wave, for interpretation by the processing device or for providing instructions or data to the processing device. The software may also be distributed over networked computer systems and stored or executed in a distributed manner. The software and data may be stored on one or more computer-readable recording media.

[0086] The above description is merely an illustrative illustration of the technical idea of ​​the present invention, and those skilled in the art will appreciate that various modifications and variations can be made without departing from the essential quality of the present invention. Therefore, the embodiments disclosed in this specification are not intended to limit the technical idea of ​​the present invention, but rather to illustrate it, and the scope of the technical idea of ​​the present invention is not limited by these embodiments. The scope of protection of the present invention should be interpreted by the following claims, and all technical ideas within a scope equivalent thereto should be interpreted as being included in the scope of the rights of the present invention.

Claims

1. A method for performing a work risk assessment recommendation device, A step of receiving a worker text, which is at least one text corresponding to a task to be performed by the worker; A step of standardizing the above worker text into a standard text for job risk assessment; generating a candidate list including at least one work risk assessment format corresponding to the above standard text; and A step of generating a recommendation list, which is at least one task risk assessment format to be recommended to the user from the candidate list based on a similarity analysis of the standard text; Recommended methods for assessing occupational risks.

2. In paragraph 1, The above worker text is, Contains text corresponding to at least one of the following: work location, worker affiliation, work target, tool utilized, and work type. Recommended methods for assessing occupational risks.

3. In paragraph 1, The steps to standardize to the above standard text are: By retrieving at least one text corresponding to each text included in said worker text from a database containing multiple standard texts, Recommended methods for assessing occupational risks.

4. In paragraph 3, The steps to standardize to the above standard text are: If a text corresponding to the text included in the above worker text does not exist in the above database, a registration request for a text that does not exist in the above database is generated. Recommended methods for assessing occupational risks.

5. In paragraph 1, The steps to standardize to the above standard text are: Further comprising extracting a superordinate concept text of the standard text based on the standard text. Recommended methods for assessing occupational risks.

6. In paragraph 5, The steps for generating the above candidate list are: By searching for at least one work risk assessment format corresponding to the above upper concept text in a database including a plurality of work risk assessments, thereby generating the candidate list, Recommended methods for assessing occupational risks.

7. In paragraph 1, The above standard text is, Includes at least one of the following: work location, self-employed affiliation, work target, tools used, and type of work; The above upper concept text is, Including at least one of the upper concept of the above work target, the upper concept of the above utilization tool, and the upper concept of the above work type, Recommended methods for assessing occupational risks.

8. In paragraph 6, The steps for generating the above candidate list are: Generating the candidate list by converting the at least one task risk assessment format corresponding to the above upper concept text into a word form or sentence form of the above upper concept text. Recommended methods for assessing occupational risks.

9. Memory containing instructions; and A processor that performs a method including: receiving a worker text, which is at least one text corresponding to a task to be performed by a worker by executing the above command; standardizing the worker text into a standard text, which is a text for task risk assessment; generating a candidate list including at least one task risk assessment format corresponding to the standard text; and generating a recommendation list, which is at least one task risk assessment format to be recommended to the user, from among the candidate list based on a similarity analysis of the standard text. Recommended device for occupational risk assessment.

10. In paragraph 9, The above worker text is, Contains text corresponding to at least one of the following: work location, worker affiliation, work target, tool utilized, and work type. Recommended device for occupational risk assessment.

11. In paragraph 9, The steps to standardize to the above standard text are: By retrieving at least one text corresponding to each text included in said worker text from a database containing multiple standard texts, Recommended device for occupational risk assessment.

12. In paragraph 11, The steps to standardize to the above standard text are: If a text corresponding to the text included in the above worker text does not exist in the above database, a registration request for a text that does not exist in the above database is generated. Recommended device for occupational risk assessment.

13. In paragraph 9, The steps to standardize to the above standard text are: Further comprising extracting a superordinate concept text of the standard text based on the standard text. Recommended device for occupational risk assessment.

14. In paragraph 13, The steps for generating the above candidate list are: By searching for at least one work risk assessment format corresponding to the above upper concept text in a database including a plurality of work risk assessments, thereby generating the candidate list, Recommended device for occupational risk assessment.

15. In paragraph 9, The above standard text is, Includes at least one of the following: work location, self-employed affiliation, work target, tools used, and type of work; The above upper concept text is, Including at least one of the upper concept of the above work target, the upper concept of the above utilization tool, and the upper concept of the above work type, Recommended device for occupational risk assessment.

16. In paragraph 14, The steps for generating the above candidate list are: Generating the candidate list by converting the at least one task risk assessment format corresponding to the above upper concept text into a word form or sentence form of the above upper concept text. Recommended device for occupational risk assessment.

17. A non-transitory computer-readable recording medium storing a computer program including instructions for causing a processor to perform a work risk assessment recommendation method, the method comprising: a step of receiving at least one worker text corresponding to a task to be performed by a worker; a step of standardizing the worker text into a standard text for work risk assessment; a step of generating a candidate list including at least one work risk assessment format corresponding to the standard text; and a step of generating a recommendation list including at least one work risk assessment format to be recommended to the user from the candidate list based on a similarity analysis of the standard text.

Images