JP2021163089A - Skill degree evaluation system - Google Patents

Abstract

To provide a skill degree evaluation system which can objectively evaluate a skill degree of work by a worker.SOLUTION: In a skill degree evaluation system, a skill degree evaluation device 5 comprises a microphone array 3 arranged in a working area in which work accompanied by the generation of a working sound is performed by a plurality of workers, and evaluates a skill degree of the worker on the basis of sound data which are acquired by the microphone array 3. The skill degree evaluation device comprises: a working sound data creation part 51 for specifying a generation source position of the working sound on the basis of the sound data, extracting the working sound generated from the generation source from the sound data, and creating working sound data; a worker specification part 54 for making the generation source position which is specified by the working sound data creation part 51 and an evaluation objective worker associated with each other; a work kind information acquisition part 55 for acquiring work kind information being information of a work kind which is performed by the evaluation objective worker; and a skill degree evaluation part 56 for evaluating a skill degree of the evaluation objective worker on the basis of the work kind information and the working sound data for each work kind.SELECTED DRAWING: Figure 2

Description

The present invention relates to a proficiency evaluation system. More specifically, the present invention relates to a proficiency evaluation system that evaluates the proficiency level of an operator based on sound data acquired by a microphone array provided in a work area.

In recent years, automation by robots is progressing in factories. For this reason, the difficulty level of the work performed by the worker tends to increase, and the worker is required to have specialized skills and sufficient proficiency. In addition, workers are required to have multiple skills. Therefore, if a plurality of workers are not arranged in appropriate work areas according to their proficiency levels, the work efficiency of each is lowered, and the overall production efficiency is also lowered.

Japanese Unexamined Patent Publication No. 2000-331069

Patent Document 1 discloses a technique for optimally arranging workers based on a worker database including desired work data and work skill level data. However, Patent Document 1 does not fully study a method for objectively evaluating the worker skill level of a worker.

An object of the present invention is to provide a proficiency evaluation system capable of objectively evaluating the proficiency level of work by an operator.

(1) The proficiency level evaluation system according to the present invention (for example, the proficiency level evaluation system 1, 1A described later) is a work area (for example, a work area described later) in which work accompanied by a work sound is performed by a plurality of workers. The microphone array provided in A) (for example, the microphone array 3 described later) is provided, and the proficiency level of the operator is evaluated based on the sound data acquired by the microphone array. The proficiency evaluation system identifies the position of the source of the work sound based on the sound data, extracts the work sound generated from the source from the sound data, and generates the work sound data. A worker identification unit (for example, a worker identification unit 54 described later) that associates a unit (for example, a work sound data generation unit 51 described later) with a position of a source specified by the work sound data generation unit and an evaluation target person. ), The work type information acquisition unit (for example, the work type information acquisition unit 55 described later) that acquires the work type information that is the work type information performed by the evaluation target person, the work type information, and the work sound. It is characterized by including a proficiency level evaluation unit (for example, a proficiency level evaluation unit 56, 56A described later) that evaluates the proficiency level of the evaluation target person for each work type based on the data.

(2) In this case, the proficiency level evaluation system stores the expert work sound data generated by recording the work sound generated by the expert performing the work in a state associated with the work type. A medium (for example, a storage medium 57 described later) is provided, and the proficiency level evaluation unit compares the work sound data with the expert work sound data of the work type specified based on the work type information. Therefore, it is preferable to evaluate the proficiency level of the evaluation target person for each work type.

(3) In this case, the proficiency level evaluation system includes a storage medium (for example, a storage medium 57A described later) that stores and stores the work sound data in a state associated with the work type and the recording date and time, and the proficiency level. It is preferable that the degree evaluation unit evaluates the proficiency level of the evaluation target person for each work type by evaluating the change with time of the plurality of the work sound data stored in the storage medium for each work type.

(4) In this case, the proficiency level evaluation unit determines the work sound generation time, the work sound interval, the work sound volume level, the work sound frequency component, the abnormal sound generation frequency, and the difference calculated from the work sound data. It is preferable to evaluate the degree of proficiency based on at least one change over time in the number of sound generations.

(1) In the present invention, the work sound data generation unit identifies the position of the work sound source in the work area based on the sound data acquired by the microphone array, and works generated from the sound data from the source. Sound is extracted, work sound data is generated, the worker identification unit associates the position of the source specified by the worker data creation department with the evaluation target person, and the work type information acquisition department is the evaluation target person. The proficiency evaluation unit acquires the work type information, which is the information of the work type being performed, and the proficiency evaluation unit is based on the work type information of the evaluation target person and the work sound data including the work sound by the evaluation target person, and this evaluation target person. Evaluate the degree of proficiency for each work type. As described above, in the present invention, the proficiency level of the work by the target worker can be objectively evaluated by using the data related to the sound generated during the work. Further, in the present invention, the position of the source of the work sound is specified from the sound data that can include the work sound by a plurality of workers, the work sound generated from this source is extracted as the work sound data, and the work sound data is further specified. By associating the location of the source of can. Therefore, it is possible to evaluate each proficiency level in real time while a plurality of workers are working in the work area.

(2) In the present invention, the proficiency level evaluation unit evaluates the proficiency level of the evaluation target person by comparing the work sound data with the expert work sound data of the work type specified based on the worker type information. do. This makes it possible to objectively evaluate the proficiency level of the worker based on the skilled person.

(3) In the present invention, the proficiency level evaluation unit evaluates the proficiency level of the evaluation target person by evaluating the change with time of the work sound data for each work type. As a result, the proficiency level of the worker can be objectively evaluated based on the initial worker himself / herself.

(4) In the present invention, the proficiency level evaluation unit uses the work sound generation time, the work sound interval, the work sound volume level, the work sound frequency component, the abnormal sound generation frequency, and the abnormal sound calculated from the work sound data. Proficiency is evaluated based on at least one of the number of occurrences over time. This makes it possible to objectively evaluate the proficiency level of the worker.

It is a figure which shows the structure of the proficiency level evaluation system which concerns on 1st Embodiment of this invention. It is a figure which shows the structure of the proficiency level evaluation apparatus. It is a figure which shows the structure of the proficiency evaluation apparatus of the proficiency evaluation system which concerns on 2nd Embodiment of this invention.

<First Embodiment>
Hereinafter, the proficiency level evaluation system according to the first embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a diagram showing a configuration of a proficiency level evaluation system 1. The proficiency level evaluation system 1 includes a microphone array 3 provided in the work area A in the factory, and a plurality of workers 2a engaged in work in the work area A based on the sound data acquired by the microphone array 3. It is provided with a proficiency level evaluation device 5 for evaluating the proficiency level of 2b and 2c.

In the work area A, a plurality of workers (3 in the example of FIG. 1) 2a, 2b, and 2c perform work of the work type defined for each of the work W conveyed by the production line L. In the following, the work W is used as the vehicle body, and the three workers 2a, 2b, and 2c each generate a work sound for the work W, and more specifically, a tightening work using an electric screwdriver for the work W. However, the present invention is not limited to this.

The microphone array 3 includes a plurality of microphones. The microphone array 3 is provided in the work area A, more specifically, at a position far from the work W so that the work sound generated by the tightening work by the workers 2a, 2b, and 2c can be recorded. It is preferable that the microphone array 3 is provided at a position where the directions to the workers 2a, 2b, and 2c are different for many hours. In the present embodiment, a case where a spherical microphone array in which a plurality of microphones are arranged in a spherical shape is used as the microphone array 3 will be described, but the present invention is not limited to this. The sound data obtained by the microphone array 3 is transmitted to the proficiency level evaluation device 5.

FIG. 2 is a diagram showing the configuration of the proficiency level evaluation device 5. The proficiency level evaluation device 5 extracts the work sound of each worker 2a, 2b, 2c from the sound data acquired by the microphone array 3 while each worker 2a, 2b, 2c is performing the work, and performs the work. The work sound data generation unit 51 that generates sound data, the evaluation unit 52 that evaluates the proficiency level of each worker 2a, 2b, 2c based on the generated work sound data, and the evaluation result by the evaluation unit 52 are displayed. A display panel 53 is provided.

The work sound data generation unit 51 has a sound source localization function that estimates the position of the sound source (more specifically, the direction from the installation position of the microphone array 3 to the sound source) from the sound data acquired by the microphone array 3. A computer equipped with software (for example, HARK (Honda Research Institute Japan Audition for Robots with Kyoto University)) having a sound source extraction function for extracting only the sound emitted from the sound source at a specific position from the sound data is used.

First, the work sound data generation unit 51 uses a predetermined sound source localization algorithm (for example, MUSIC (multiple signal classification) method) for sound data including work sounds generated by tightening work by each worker 2a, 2b, 2c. ) Based on the sound source extraction process, the position of the source of the work sound (more specifically, the direction from the installation position of the microphone array 3 to the source of the work sound) is specified. According to the sound source localization algorithm, even when the workers 2a, 2b, and 2c perform the tightening work at the same time, that is, the sound data includes work sounds generated from a plurality of different source positions at the same time. Even if this is the case, the location of the source of these work sounds can be specified.

Next, the work sound data generation unit 51 uses a predetermined sound source extraction algorithm (for example, GHDSS (Geometric High-order Diskorrelation base Separation) method) for the sound data for which the source position of the work sound is specified as described above. By performing the sound source extraction process based on the above, only the sound generated from the source position of the work sound specified as described above is extracted from the original sound data, and the work sound data is generated. Therefore, in the work sound data, among various sounds included in the original sound data, sounds other than the sound generated from the specified source position are suppressed. Further, when the work sound data generation unit 51 simultaneously includes work sounds generated from a plurality of different source positions as described above, the work sound data generation unit 51 specifies each source position based on the sound source localization algorithm. Generate work sound data.

The work sound data generation unit 51 evaluates the work sound data generated from the original sound data as described above in a state of being associated with the information related to the source position (hereinafter, also referred to as “source position information”). Send to 52.

The evaluation unit 52 includes a worker identification unit 54, a work type information acquisition unit 55, a proficiency level evaluation unit 56, and a storage medium 57.

The worker identification unit 54 associates the source position information transmitted from the work sound data generation unit 51 with the evaluation target person who is one of the plurality of workers 2a, 2b, and 2c. More specifically, the worker identification unit 54 identifies the worker who generated the work sound included in the work sound data associated with the source position information as the evaluation target person based on the source position information. do. The worker identification unit 54 identifies the person to be evaluated based on the image taken by the camera provided in the factory and the source position information, but the present invention is not limited to this. When the movement range of each worker 2a, 2b, 2c is limited, the evaluation target person can be specified only from the source position information.

The work type information acquisition unit 55 has a worker number unique to the evaluation target person specified by the worker identification unit 54 and the work currently being performed by the evaluation target person (when sound data is acquired by the microphone array 3). The work type information, which is the information of the type (hereinafter referred to as "evaluation target work type"), is acquired, and the information about these evaluation target persons is transmitted to the proficiency level evaluation unit 56 as the evaluation target person information.

The storage medium 57 stores a plurality of expert work sound data created in advance for each work type. The expert work sound data is generated by recording the work sound generated by the expert performing the work for each work type. For example, in tightening work using an electric screwdriver, the work sound generated when a skilled worker performs this tightening work is the work sound generated when a beginner who is not accustomed to this tightening work performs the tightening work. different. For example, when a skilled person performs the tightening work, almost a constant level of striking sound is generated at approximately equal intervals, whereas when a beginner performs the tightening work, the level of the striking sound generated is not constant. In addition, the intervals between the striking sounds are also different, and abnormal sounds are also generated. In the storage medium 57, such expert work sound data is stored in a state associated with the work type.

The proficiency level evaluation unit 56 evaluates the work sound data generated by the work sound data generation unit 51, and the evaluation target person associated with the work sound data by the worker identification unit 54 and the work type information acquisition unit 55. Based on the information, the proficiency level of the evaluation target person is evaluated for each work type. More specifically, the proficiency level evaluation unit 56 reads the expert work sound data associated with the work type to be evaluated from the storage medium 57, and also reads the work sound data generated by the work sound data generation unit 51. However, by comparing the expert work sound data with a known algorithm, the proficiency level of the work of the work type to be evaluated by the evaluation target person is evaluated. The proficiency level evaluation unit 56 evaluates the proficiency level higher as the degree of matching between the work sound data and the expert work sound data increases, and lowers the proficiency level as the degree of matching between the work sound data and the expert work sound data decreases. evaluate.

The proficiency level evaluation unit 56 stores the proficiency level evaluation result in the storage medium 57 for each worker and work type. Further, the evaluation result stored in the storage medium 57 can be appropriately displayed on the display panel 53 according to the operation by the administrator. Therefore, by referring to this evaluation result, the manager can arrange a plurality of workers in a work area suitable for each proficiency level and improve the overall production efficiency.

According to the proficiency level evaluation system 1 according to the present embodiment, the following effects are obtained.
(1) The work sound data generation unit 51 identifies the source position of the work sound in the work area A based on the sound data acquired by the microphone array 3, and generates the work sound generated from the source from the sound data. Extract and generate work sound data. The worker identification unit 54 associates the source position specified by the worker data creation unit 51 with the evaluation target person, and the work type information acquisition unit 55 is information on the work type currently being performed by the evaluation target person. Acquire work type information. The proficiency level evaluation unit 56 evaluates the proficiency level of the evaluation target person for each work type based on the work type information of the evaluation target person and the work sound data including the work sound by the evaluation target person. In this way, the proficiency level evaluation system 1 can objectively evaluate the proficiency level of the work by the target worker by using the data related to the sound generated during the work. Further, in the proficiency evaluation system 1, the position of the source of the work sound is specified from the sound data that can include the work sound by a plurality of workers, the work sound generated from this source is extracted as the work sound data, and this work is further performed. By associating the sound data source position with the evaluation target person, even if a plurality of workers 2a, 2b, and 2c are performing work to generate work sound at the same time, only the evaluation target worker. Can properly evaluate the proficiency level of. Therefore, it is possible to evaluate each proficiency level in real time while a plurality of workers 2a, 2b, and 2c are working in the work area A.

(2) In the proficiency level evaluation system 1, the proficiency level evaluation unit 56 compares the work sound data with the expert work sound data of the work type specified based on the worker type information to evaluate the evaluation target person. Evaluate the degree of proficiency. This makes it possible to objectively evaluate the proficiency level of the worker based on the skilled person.

<Second Embodiment>
Hereinafter, the proficiency level evaluation system according to the second embodiment of the present invention will be described with reference to the drawings. In the following description, the same components as those in the first embodiment are designated by the same reference numerals, and detailed description thereof will be omitted.

FIG. 3 is a diagram showing a configuration of a proficiency level evaluation device 5A of the proficiency level evaluation system 1A according to the present embodiment. The proficiency evaluation system 1A according to the present embodiment differs in the configuration of the evaluation unit 52A, more specifically, the configuration of the proficiency evaluation unit 56A and the storage medium 57A.

The proficiency level evaluation unit 56A evaluates the work sound data generated by the work sound data generation unit 51 with respect to the evaluation target person in which the work sound data is associated with the worker identification unit 54 and the work type information acquisition unit 55. The information (worker number and evaluation target work type) and work sound data are stored and stored in the storage medium 57A in a state associated with the recording date and time. As a result, the storage medium 57A is constructed with a work sound database which is a set of time-series data of a plurality of work sound data classified according to the worker and the work type.

Further, the proficiency level evaluation unit 56A reads the time-series data of the work sound data of the evaluation target person by appropriately referring to the work sound database constructed in the storage medium 57A, and evaluates the time-series data for each work type. The degree of proficiency of the evaluation target person is evaluated for each work type. For example, when a beginner performs tightening work using an electric screwdriver, the level of the striking sound generated is not constant, the intervals between the striking sounds are different, and abnormal noise is also generated. On the other hand, as the proficiency level increases, the level and interval of the striking sounds approach constant, and the frequency of occurrence of abnormal sounds is considered to decrease. Therefore, the proficiency level evaluation unit 56A at least the work sound generation time, the work sound interval, the work sound volume level, the work sound frequency component, the abnormal sound generation frequency, and the abnormal sound generation frequency calculated from each work sound data. The degree of proficiency is evaluated based on any change over time.

The proficiency level evaluation unit 56 stores the proficiency level evaluation result in the storage medium 57A for each worker and work type. Further, the evaluation result stored in the storage medium 57A can be appropriately displayed on the display panel 53 according to the operation by the administrator. Therefore, by referring to this evaluation result, the manager can arrange a plurality of workers in a work area suitable for each proficiency level and improve the overall production efficiency.

According to the proficiency level evaluation system 1A according to the present embodiment, the following effects are obtained.
(3) The proficiency level evaluation unit 56A evaluates the proficiency level of the evaluation target person by evaluating the change with time of the work sound data for each work type. As a result, the proficiency level of the worker can be objectively evaluated based on the initial worker himself / herself.

(4) The proficiency level evaluation unit 56A determines the work sound generation time, the work sound interval, the work sound volume level, the work sound frequency component, the abnormal sound occurrence frequency, and the abnormal sound generation frequency calculated from the work sound data. The proficiency level is evaluated based on at least one of the changes over time. This makes it possible to objectively evaluate the proficiency level of the worker.

Although one embodiment of the present invention has been described above, the present invention is not limited to this. Within the scope of the gist of the present invention, the detailed configuration may be changed as appropriate.

1,1A ... Proficiency evaluation system A ... Work area W ... Work 2a, 2b, 2c ... Worker 3 ... Microphone array 5,5A ... Proficiency evaluation device 51 ... Work sound data generation unit 52, 52A ... Evaluation unit 53 ... Display panel 54 ... Worker identification unit 55 ... Work type information acquisition unit 56, 56A ... Proficiency evaluation unit 57, 57A ... Storage medium

Claims (4)

A proficiency evaluation system that evaluates the proficiency level of workers based on the sound data acquired by the microphone array, which is equipped with a microphone array provided in a work area where work involving the generation of work sounds is performed by a plurality of workers. And
A work sound data generation unit that identifies the position of the work sound generation source based on the sound data, extracts the work sound generated from the source from the sound data, and generates the work sound data.
A worker identification unit that associates the position of the source identified by the work sound data generation unit with the evaluation target person,
A work type information acquisition unit that acquires work type information, which is work type information performed by the evaluation target person, and a work type information acquisition unit.
A proficiency level evaluation system including a proficiency level evaluation unit that evaluates the proficiency level of the evaluation target person for each work type based on the work type information and the work sound data. It is provided with a storage medium that stores the expert work sound data generated by recording the work sound generated by the expert performing the work in a state associated with the work type.
The proficiency level evaluation unit compares the work sound data with the expert work sound data of the work type specified based on the work type information to determine the proficiency level of the evaluation target person for each work type. The proficiency level evaluation system according to claim 1, wherein the proficiency level is evaluated. A storage medium for storing and accumulating the work sound data in a state associated with the work type and the recording date and time is provided.
The proficiency level evaluation unit is characterized in that the proficiency level of the evaluation target person is evaluated for each work type by evaluating the time course of a plurality of the work sound data stored in the storage medium for each work type. The proficiency evaluation system according to claim 1. The proficiency level evaluation unit is at least any one of the work sound generation time, the work sound interval, the work sound volume level, the work sound frequency component, the abnormal sound generation frequency, and the abnormal sound generation frequency calculated from the work sound data. The proficiency level evaluation system according to claim 3, wherein the proficiency level is evaluated based on the change over time.

Images