JP2020095651A - Productivity evaluation system, productivity evaluation device, productivity evaluation method, and program - Google Patents

Abstract

To provide a productivity evaluation system for evaluating an operation rate of a manufacturing facility at a relatively low introduction cost.SOLUTION: A productivity evaluation system includes: a camera installed for photographing a manufacturing site; a position identification part which identifies a position of a human from among images photographed by the camera; and an evaluation part which generates an evaluation value for evaluating an operation rate of the manufacturing facility installed in the manufacturing site. Preferably further included is a human extraction part which extracts a feature image of the human from among the images photographed by the camera, the position identification part identifies a position of the feature image of the human extracted by the human extraction part, and the evaluation part determines whether or not the feature image of the human is present in a region set with reference to at least one of an installation position and an installation direction of the manufacturing facility and generates the evaluation value based on the number of people present in the set region.SELECTED DRAWING: Figure 2

Description

The present invention relates to a productivity evaluation system, a productivity evaluation device, a productivity evaluation method, and a program.

For example, in Patent Document 1, there is a work quality control method for a worker with respect to a work in which repetitive motions are performed. The flow line data of the past worker, past intermediate quality data of the work target product, the work target Model pasting step 102 for statistically constructing a predictive model in which the flow line data and the intermediate quality data are input and the final quality data is output from the past final quality data of the product; Substituting the worker position recognition step 104 for recognizing the position of the worker and the worker position recognized in the worker position recognition step into the model constructed in the model construction step to determine whether the target work is normal. There is disclosed a work quality control method including a worker position abnormality determination step 105 for determining whether there is an abnormality.

Patent Document 2 discloses a body condition of an operator at each point in time in which a worker performing a work to be measured is simply modeled by a detection point corresponding to a plurality of body parts and a line connecting the detection points. Detection information acquisition unit that acquires information of the operator, an evaluation value calculation unit that calculates an evaluation value for the worker's work based on the information obtained by the detection information acquisition unit, and a knitting content that is different for a group of processes to be knitted A knitting simulation for obtaining a plurality of process knitting is performed, and one or more process knitting is selected from the plurality of process knitting based on the evaluation value for each process knitting obtained by the knitting simulation, and the selected process knitting is selected. And an output control unit that performs control for presenting the information of 1.
In Patent Document 3, when the tracking target area is selected by the user from the frame displayed on the monitor 500, the area is acquired as tracking target information and stored in the tracking target information storage unit 354. The tracking target object is tracked based on the tracking target information stored in the unit 354 and the probability distribution image generated by a predetermined tracking method, and the tracking result data is stored in the tracking result data storage unit 356. The color information is associated with each frame of the tracking result data stored in 356, a line is drawn from the center coordinate of the frame located immediately before in time to the center coordinate of the next frame, and this process is performed. There is disclosed a trajectory drawing system which is repeatedly executed until it cannot be obtained.

JP, 2016-157357, A JP, 2017-068432, A JP, 2016-042306, A

It is an object of the present invention to provide a productivity evaluation system that evaluates the operating rate of manufacturing equipment at a relatively low introduction cost.

The productivity evaluation system according to the present invention includes a camera installed to photograph a manufacturing site, a position specifying unit that specifies a position of a person from an image taken by the camera, and the position specifying unit. And an evaluation unit that generates an evaluation value for evaluating the operating rate of the manufacturing equipment installed at the manufacturing site based on the specified position of the person.

Suitably, it further has a person extraction unit that extracts a characteristic image of a person from the images taken by the camera, and the position specifying unit is a position of the characteristic image of the person extracted by the person extraction unit. The evaluation unit determines whether or not a characteristic image of a person exists in the area set based on at least one of the installation position and the installation direction of the manufacturing facility, and exists in the set area. The evaluation value is generated based on the number of people who perform the evaluation.

Preferably, the camera captures a plurality of still images or moving images that are temporally continuous, and the evaluation unit calculates the staying time and the number of people existing in the set area as the evaluation value. To do.

Suitably, it further has a display part which superimposes and displays the evaluation value image which shows the evaluation value generated by the above-mentioned evaluation part on one of the pictures picturized by the above-mentioned camera.

Preferably, the operating rate of the manufacturing facility is evaluated based on the voice at the manufacturing site or the detection unit that detects the power consumed by the manufacturing facility, and the voice or the power detected by the detection unit. A second evaluation unit for generating a second evaluation value for the purpose, a staying time or the number of people calculated by the evaluation unit, and a second evaluation value generated by the second evaluation unit are arranged so that they can be compared with each other. And a display unit for displaying the information.

Preferably, a scanner device that scans the business report described by the worker at the manufacturing site, a business report scanned by the scanner device, a description extraction unit that extracts the description items, and based on the time axis, The description further includes a display unit that displays the description items extracted by the description extraction unit and the stay time or the number of persons calculated by the evaluation unit in association with each other.

Further, the productivity evaluation apparatus according to the present invention is a position specifying unit that specifies a position of a person from among images taken by a camera installed at a manufacturing site, and a position of the person specified by the position specifying unit. And an evaluation unit that generates an evaluation value for evaluating the operating rate of the manufacturing facility installed at the manufacturing site.

Further, in the productivity evaluation method according to the present invention, a camera takes a picture of a manufacturing site, a computer specifies a position of a person from an image taken by the camera, and a computer specifies Generating an evaluation value for evaluating the operating rate of the manufacturing equipment installed at the manufacturing site based on the position of the person who has performed the operation.

Further, the program according to the present invention is based on the step of identifying the position of a person from the images captured by the camera installed to capture the production site, and the manufacturing based on the identified position of the person. And a step of generating an evaluation value for evaluating the operating rate of the manufacturing equipment installed at the site.

The operating rate of manufacturing equipment can be evaluated at a relatively low introduction cost.

It is a figure which illustrates the whole structure of the productivity evaluation system 1. It is a figure which illustrates the manufacturing equipment 900 installed in the manufacturing factory 9, and the work area 902 of a person. It is a figure which illustrates the hardware constitutions of the productivity evaluation apparatus 2. It is a figure which illustrates the functional structure of the productivity evaluation apparatus 2. It is a figure explaining the evaluation part 230 in detail. It is a flow chart explaining camera control processing (S10) by tablet terminal 4. It is a flow chart explaining flow line visualization processing (S20) by productivity evaluation device 2. It is a figure which illustrates the graph showing the number of persons staying per unit time in the work area 902 of the manufacturing equipment 900. It is a display example in which the analysis result of the business report and the evaluation result by the captured image are displayed in a comparable manner. It is a figure which illustrates the heat map which shows the stay time of the worker in each area|region in the manufacturing factory 9 by color. It is a figure which illustrates the functional structure of the productivity evaluation system 1 in the modification 1. 9 is a flowchart illustrating a camera control process (S12) of the tablet terminal 4 according to Modification 1. 9 is a flowchart illustrating a flow line visualization process (S22) of the productivity evaluation device 2 in Modification Example 1. It is a figure which illustrates the functional structure of the productivity evaluation apparatus 2 in the modification 2. (A) is a figure explaining the 2nd evaluation part 250 using a voice in more detail, (B) is a figure explaining the 2nd evaluation part 250 using a power value in more detail. It is a figure which illustrates the structure of the 2nd evaluation part 250 which compares work content. It is a flow chart explaining work contents comparison processing (S30).

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a diagram illustrating an overall configuration of the productivity evaluation system 1.
As illustrated in FIG. 1, the productivity evaluation system 1 includes a productivity evaluation device 2 and a tablet terminal 4, which are connected to each other via a network 7. A scanner device 3 is connected to the productivity evaluation device 2.

The productivity evaluation device 2 is an example of the productivity evaluation device according to the present invention, and is a computer terminal on which a productivity evaluation program 22 (described later) operates. In this example, the productivity evaluation device 2 which is a stand-alone computer terminal is described as a specific example, but the invention is not limited to this, and may be, for example, a virtual machine on a cloud.
The productivity evaluation device 2 of this example generates evaluation information for evaluating the operating rate of the manufacturing facility 900 (described later) based on the image captured by the tablet terminal 4, and displays the generated evaluation information. ..
The scanner device 3 scans the business report written by the worker at the manufacturing site.

The tablet terminal 4 is a computer terminal with a built-in camera and is an example of the camera according to the present invention. The tablet terminal 4 photographs the manufacturing site where the manufacturing facility 900 is installed with a built-in camera, and transmits the photographed image data to the productivity evaluation device 2. For example, the tablet terminal 4 is installed at a position where a plurality of manufacturing facilities 900 installed in a manufacturing factory can be captured, and captures a plurality of temporally continuous still images or moving images. The tablet terminal 4 of this example is installed at a position where the entire manufacturing factory 9 can be photographed, photographs a moving image, and transmits the photographed moving image data to the productivity evaluation device 2.

FIG. 2 is a diagram illustrating a manufacturing facility 900 installed in the manufacturing factory 9 and a work area 902 for a person.
As illustrated in FIG. 2, the manufacturing facility 900 includes, for example, a drilling machine 900A, a laser processing machine 900B, and a press processing machine 900C. In each of these manufacturing facilities 900, a work area 902 in which a worker will work is defined based on the installation position and installation direction (installation direction).
The tablet terminal 4 of this example is installed at a position where at least the work area 902 can be photographed, and the productivity evaluation device 2 determines the number of workers existing in the work area 902 based on the image photographed by the tablet terminal 4. And the staying time are calculated, and the evaluation information for evaluating the operating rate of the manufacturing facility 900 is generated based on the calculated number of people and the staying time.
There is an analog device such as a drilling machine 900A in the manufacturing facility 900, and network connection is not possible. Therefore, in the productivity evaluation system 1 of this example, the operating rate of the manufacturing facility 900 is indirectly evaluated from the number of workers in the work area 902 and the staying time.
Further, in the productivity evaluation system 1 of the present example, in addition to the work area 902 of the manufacturing facility 900, the passage that is the flow line of the worker, the material storage 910, and the like are also photographed, and the staying time of the worker in these places is taken. By also calculating, it is possible to extract the process that is the bottleneck in manufacturing, and evaluate the productivity of the entire manufacturing factory 9.

FIG. 3 is a diagram illustrating a hardware configuration of the productivity evaluation device 2.
As illustrated in FIG. 3, the productivity evaluation device 2 includes a CPU 200, a memory 202, an HDD 204, a network interface 206 (network IF 206), a display device 208, and an input device 210, and these configurations are via a bus 212. Connected to each other.
The CPU 200 is, for example, a central processing unit.
The memory 202 is, for example, a volatile memory and functions as a main storage device.
The HDD 204 is, for example, a hard disk drive device, and stores a computer program (for example, the productivity evaluation program 22 in FIG. 4) and other data files as a nonvolatile recording device.
The network IF 206 is an interface for wired or wireless communication.
The display device 208 is, for example, a liquid crystal display.
The input device 210 is, for example, a keyboard and a mouse.

FIG. 4 is a diagram illustrating a functional configuration of the productivity evaluation device 2.
As illustrated in FIG. 4, the productivity evaluation program 22 is installed in the productivity evaluation device 2, and the captured image database 260 (captured image DB 260), the business report database 270 (business report DB 270), and the evaluation result database 280. (Evaluation result DB 280) is configured. The productivity evaluation program 22 is stored in, for example, a recording medium such as a CD-ROM, and is installed in the productivity evaluation device 2 via this recording medium.
Note that part or all of the productivity evaluation program 22 may be realized by hardware such as an ASIC, or may be realized by partially borrowing a function of an OS (Operating System). Further, all of these programs may be installed in one computer terminal or may be installed in a virtual machine on the cloud.

The productivity evaluation program 22 has an image acquisition unit 220, a skeleton detection unit 222, a skeleton coordinate extraction unit 224, a description extraction unit 226, an association unit 228, an evaluation unit 230, and a display unit 240.
In the productivity evaluation program 22, the image acquisition unit 220 acquires the image data captured at the manufacturing site. The image acquisition unit 220 of this example receives the moving image data shot by the tablet terminal 4 from the tablet terminal 4, and registers the received moving image data in the shot image DB 260 in association with the shooting date and time.

The skeleton detection unit 222 detects an image corresponding to a human skeleton from the image data acquired by the image acquisition unit 220. The skeleton detection unit 222 is an example of the human extraction unit according to the present invention, and may detect a human face as a human characteristic image instead of the image corresponding to the human skeleton.

The skeleton coordinate extraction unit 224 extracts the coordinates of the skeleton image of the person detected by the skeleton detection unit 222 as the position information of the person. The skeleton coordinate extraction unit 224 of this example extracts the coordinates of the human neck. The skeleton coordinate extracting unit 224 is an example of the position specifying unit according to the present invention.

The description extraction unit 226 extracts description items from the business report scanned by the scanner device 3. For example, the description extraction unit 226 extracts the work content and the work execution date and time from the scanned business report according to the format definition of the business report. The description extraction unit 226 of this example extracts the work content and the work execution date and time from the scanned work report, and registers the extracted work content and the execution date and time in the work report DB 270 in association with each other.

The associating unit 228 associates the description items extracted by the description extracting unit 226 with the evaluation result by the evaluation unit 230 based on the time axis. For example, the associating unit 228 compares the work content and the implementation date and time described in the daily report with the stay time or the number of guests in each work area 902 calculated by the evaluation unit 230, and associates each other.

The evaluation unit 230 generates an evaluation value for evaluating the operating rate of the manufacturing facility 900 based on the coordinates extracted by the skeleton coordinate extraction unit 224. For example, the evaluation unit 230 determines whether or not the skeleton coordinates of a person exist in each work area 902 set based on the installation position and the installation direction of the manufacturing equipment 900, and the work area 902 is set in each set work area 902. The number of existing people is counted for each time period, an evaluation value for evaluating the operating rate of the manufacturing facility 900 is generated based on the counted number of people, and the evaluation value is registered in the evaluation result DB 280. The registered evaluation value may be, for example, the number of people staying in each work area 902 itself, or the number of people staying normalized by the number of all workers.

More specifically, as illustrated in FIG. 5, the evaluation unit 230 includes a number-of-people calculation unit 232 and a flow line visualization unit 234. The number-of-persons calculation unit 232 determines whether or not the skeleton coordinates of the person exist in each of the work areas 902, counts the number of persons in each of the set work areas 902 for each time zone, and determines each time zone. The number of people staying at is output as an evaluation value. As illustrated in FIG. 8, the number-of-persons calculation unit 232 of this example generates a graph showing the number of persons staying per unit time in the work area 902 of each manufacturing facility 900. In the graph of FIG. 8, the horizontal axis represents the time zone and the vertical axis represents the number of people staying per unit time. In this example, the number of workers staying in the work area of each manufacturing facility 900 per unit time is an index of the operating rate of the manufacturing facility 900.

The flow line visualization unit 234 generates an image that visualizes the flow line of a person on the manufacturing site based on the coordinates extracted by the skeleton coordinate extraction unit 224. As illustrated in FIG. 10, the flow line visualization unit 234 of the present example generates a heat map that indicates the staying time of the worker in each region in the manufacturing factory 9 by color. In the heat map of FIG. 10, the number of workers staying or the staying time in each area (including the work area, the passage, the material storage 910, etc.) in the manufacturing factory 9 is represented by a color. The flow lines and the main places of stay are visualized.

The display unit 240 superimposes and displays the evaluation value image indicating the evaluation value generated by the evaluation unit 230 on any of the images captured by the tablet terminal 4. The evaluation value image is, for example, an image in which the number of people staying in each area or the staying time is represented by a color, a pattern, or the like. The display unit 240 of this example displays the heat map generated by the flow line visualization unit 234 by superimposing it on the image captured by the tablet terminal 4.

In addition, the display unit 240 displays the description items associated by the association unit 228 and the stay time or the number of persons calculated by the evaluation unit 230 in association with each other. As illustrated in FIG. 9, the display unit 240 of the present example is extracted by the description extraction unit 226 and the staying time of each work area 902 calculated by the evaluation unit 230 (that is, the operating rate of the manufacturing equipment 900). The items described in the business report (analysis results of the business report) are displayed in association with each other. The analysis result of the work report in FIG. 9 is obtained by specifying the work content of each worker from the items described in the work report and calculating the required time of the specified work content. Depending on whether or not there is a correlation between the operating rate of the manufacturing facility 900 and the time required for the work, it is possible to study the efficiency of the factory layout and facility expansion.

FIG. 6 is a flowchart illustrating the camera control process (S10) by the tablet terminal 4.
As illustrated in FIG. 6, in step 100 (S100), the tablet terminal 4 waits until the power is turned on (S100: No), and when the power is turned on, the process proceeds to step S105.

In step 105 (S105), the tablet terminal 4 starts a time timer.
In step 110 (S110), the tablet terminal 4 initializes the built-in camera.

In step 115 (S115), the tablet terminal 4 starts capturing a moving image with the built-in camera.
In step 120 (S120), the tablet terminal 4 refers to the time timer, determines whether it is during business hours, and if it is during business hours, the tablet terminal 4 shifts to the processing of S125 and is outside business hours. In some cases, the shooting of the moving image is ended, and the camera control process (S10) is completed.

In step 125 (S125), the tablet terminal 4 extracts a frame from the captured moving image data.
In step 130 (S130), the tablet terminal 4 transmits the extracted frame to the productivity evaluation device 2, and returns to the process of S120.

FIG. 7 is a flowchart illustrating the flow line visualization processing (S20) by the productivity evaluation device 2.
As illustrated in FIG. 7, in step 200 (S200), the productivity evaluation apparatus 2 waits until the power is turned on (S200: No), and when the power is turned on, the process proceeds to step S205.

In step 205 (S205), the productivity evaluation apparatus 2 determines whether it is during business hours, and if it is during business hours, shifts to the processing of S210, and if it is outside business hours, the productivity evaluation apparatus 2 moves. The line visualization processing (S20) is completed.

In step 210 (S210), the image acquisition unit 220 (FIG. 4) of the productivity evaluation device 2 receives the frame data of the captured moving image from the tablet terminal 4 and registers it in the captured image DB 260.
In step 215 (S215), the image acquisition unit 220 reads out a frame to be processed from the captured image DB 260.
In step 220 (S220), the skeleton detection unit 222 extracts the skeleton of the human body from the frames read by the image acquisition unit 220.

In step 225 (S225), the skeleton coordinate extraction unit 224 extracts the coordinates of the part of interest (neck) from the skeleton detected by the skeleton detection unit 222. The extracted coordinates are the coordinates on the captured image.
In step 230 (S230), the evaluation unit 230 additionally registers the coordinates extracted by the skeleton coordinate extraction unit 224 in the flow line analysis graph. The flow line analysis graph is a graph showing the distribution of people (cumulative stay time or cumulative number of stayers) in the captured image (that is, in the manufacturing plant). When a worker exists in the same region over a plurality of frames, the worker's stay time can be calculated.
The evaluation unit 230 compares the flow line analysis graph created as described above with a preset work area 902 (FIG. 2) to evaluate the operation rate of each manufacturing facility 900. Calculate the value.

As described above, according to the productivity evaluation system 1 of the present embodiment, the staying place and the staying time of the worker are calculated based on the image taken in the manufacturing factory 9, and the operating rate of the manufacturing facility 900 is calculated. Is used as the evaluation value. As a result, even if the manufacturing facility 900 is difficult to connect to the network, the operating rate of the manufacturing facility 900 can be evaluated. Furthermore, since it is not necessary to install a sensor for detecting the operating rate and a communication device for transmitting the detection result for each manufacturing facility 900, it is possible to evaluate the operating rate of the manufacturing facility 900 at a relatively low cost. You can
Further, according to the productivity evaluation system 1 of this example, items to be described are extracted from the work report described by the worker and displayed in a manner that can be compared with the evaluation result by the evaluation unit 230. Accordingly, the subjective evaluation result of the worker and the objective evaluation result based on the captured image can be compared to study a measure for improving productivity. Furthermore, according to the productivity evaluation system 1 of this example, the moving image data registered in the captured image DB 260 and the description items registered in the work report DB 270 can be associated with each other. You can find the video information of time more easily.

[Modification 1]
Next, a modified example of the above embodiment will be described.
FIG. 11 is a diagram illustrating a functional configuration of the productivity evaluation system 1 in the first modification. It should be noted that, of the components shown in this figure, those substantially the same as the components shown in FIG. 4 are designated by the same reference numerals.
As illustrated in FIG. 11, in the first modification, the camera side program 43 operates on the tablet terminal 4 and the productivity evaluation program 23 operates on the productivity evaluation apparatus 2.
The camera-side program 43 includes an image acquisition unit 420, a skeleton detection unit 422, a skeleton coordinate extraction unit 424, and a feature amount transmission unit 426.
The productivity evaluation program 23 has a description extraction unit 226, an association unit 228, an evaluation unit 230, and a display unit 240.

In the camera side program 43, the image acquisition unit 420 acquires image data captured by the built-in camera of the tablet terminal 4.
The skeleton detection unit 422 detects an image corresponding to a human skeleton from the image data acquired by the image acquisition unit 420.
The skeleton coordinate extraction unit 424 extracts the coordinates of the skeleton image corresponding to the neck from the image of the skeleton detected by the skeleton detection unit 422.
The feature amount transmission unit 426 transmits the coordinates extracted by the skeleton coordinate extraction unit 424 to the productivity evaluation device 2.
As described above, the tablet terminal 4 according to the modified example 1 performs the detection of the skeleton and the extraction of the coordinates of the region of interest in addition to the imaging in the manufacturing factory 9, and the extracted coordinates are used by the productivity evaluation device 2. Send, do not send captured images.

FIG. 12 is a flowchart illustrating the camera control process (S12) of the tablet terminal 4 in the first modification. Note that among the steps shown in this figure, those substantially the same as the steps shown in FIG. 6 are given the same reference numerals.
FIG. 13 is a flowchart illustrating the flow line visualization processing (S22) of the productivity evaluation device 2 in the first modification. Note that among the steps shown in this figure, those substantially the same as the steps shown in FIG. 7 are given the same reference numerals.

As shown in FIG. 12, the operation from step 100 to step 120 is the same as in the above embodiment. The moving image data captured by the tablet terminal 4 is registered in the captured image DB 460 and sequentially read by the image acquisition unit 420.
In step 140 (S140), the skeleton detection unit 422 extracts the skeleton of the human body from the frames read by the image acquisition unit 420.
In step 145 (S145), the skeleton coordinate extraction unit 424 extracts the coordinates of the part of interest (neck) from the skeleton detected by the skeleton detection unit 422.
In step 150 (S150), the feature amount transmission unit 426 transmits the coordinates extracted by the skeleton coordinate extraction unit 424 to the productivity evaluation device 2.
In step 155 (S155), the feature amount transmitting unit 426 discards the processed frame from the captured image DB 460.

As shown in FIG. 13, steps 200 to 205 are the same as those in the above embodiment.
In step 210 (S210), the evaluation unit 230 (FIG. 11) receives the coordinates of the site of interest (neck) from the tablet terminal 4.
In step 240 (S240), the evaluation unit 230 additionally registers the coordinates received from the tablet terminal 4 in the flow line analysis graph.

As described above, in Modification 1, the extraction of the coordinates of the region of interest is performed on the tablet terminal 4 side and the captured image is discarded, whereby the privacy of the worker is protected. That is, the information transmitted/received via the network 7 is the coordinates of the region of interest, does not identify the individual worker, and privacy is protected.

[Modification 2]
Next, a modified example 2 of the above embodiment will be described.
In the second modification, the tablet terminal 4 records the sound generated in the manufacturing factory 9 with the built-in microphone in addition to photographing the manufacturing factory 9. Further, the tablet terminal 4 communicates with an outlet or a power meter (smart meter) that supplies power to the manufacturing facility 900 to detect the power consumed by the manufacturing facility 900.
The productivity evaluation device 2 of the modified example 2 generates a second evaluation value different from the first evaluation value based on the captured image based on the voice detected by the tablet terminal 4 or the power consumption.

FIG. 14 is a diagram illustrating a functional configuration of the productivity evaluation device 2 in the second modification. It should be noted that, of the components shown in this figure, those substantially the same as the components shown in FIG. 4 are designated by the same reference numerals.
As illustrated in FIG. 14, the productivity evaluation program 24 of the modified example has a configuration in which a second evaluation unit 250 is added to the productivity evaluation program 22 of FIG.
The 2nd evaluation part 250 produces|generates the 2nd evaluation value for evaluating the operating rate of a manufacturing equipment based on the voice in the manufacturing factory 9, or the electric power consumed by the manufacturing equipment 900.
The display unit 240 of the present example displays the evaluation result by the second evaluation unit 250 so as to be comparable to the staying time or the number of people (first evaluation value) calculated by the evaluation unit 230.

FIG. 15(A) is a diagram illustrating the second evaluation unit 250 that uses voice in more detail, and FIG. 15(B) is a diagram that illustrates the second evaluation unit 250 that uses a power value in more detail. Is.
As illustrated in FIG. 15A, the second evaluation unit 250 that uses voice includes a recording control unit 252, a sound source separation unit 254, a device determination unit 256, and an operation rate calculation unit 258.
The recording control unit 252 controls the microphone of the tablet terminal 4 to record the sound generated inside the manufacturing factory 9.
The sound source separation unit 254 extracts the sound data of a specific sound source from the sound data recorded under the control of the sound recording control unit 252. For example, the sound source separation unit 254 extracts the sound data of each sound source based on the frequency, the arrival time to a plurality of microphones, or the like according to the sound source separation technique.
The device discriminating unit 256 discriminates the manufacturing facility 900, which is a sound generation source, based on each audio data extracted by the sound source separating unit 254. For example, the device discrimination unit 256 discriminates the drilling sound of the drilling machine 900A, the processing sound of the laser processing machine 900B, and the pressing sound of the press processing machine 900C based on the frequency included in the audio data.
The operation rate calculation unit 258 calculates the operation rate of the manufacturing facility 900 based on the determination result by the device determination unit 256 and the determined duration of the sound, and sets it as the second evaluation value.

As illustrated in FIG. 15B, the second evaluation unit 250 that uses the power value includes a power value acquisition unit 260 and an operating rate calculation unit 262.
The power value acquisition unit 260 acquires the value of the power consumed by the manufacturing facility 900. For example, the power value acquisition unit 260 may directly communicate with a power sensor installed in the manufacturing facility 900 to acquire the power consumption of each manufacturing facility 900, or via the tablet terminal 4, each manufacturing facility 900. Power consumption may be acquired.
The operating rate calculation unit 262 calculates the operating rate of the manufacturing facility 900 based on the power consumption of each manufacturing facility 900 acquired by the power value acquiring unit 260, and sets it as the second evaluation value.

The productivity evaluation device 2 of the modified example 2 calculates a second evaluation value for evaluating the operating rate of the manufacturing equipment 900 based on the voice data recorded in the manufacturing factory 9 or the power consumption of the manufacturing equipment 900, The calculated second evaluation value and the first evaluation value generated based on the captured image are displayed side by side in a comparable state. This makes it possible to evaluate the utilization efficiency of manufacturing equipment. For example, the ratio of the time required for adjustment and maintenance of the manufacturing equipment to the actual operating time can be calculated.

[Modification 3]
A further modification of Modification 2 will be described.
FIG. 16: is a figure which illustrates the structure of the 2nd evaluation part 250 which compares work content.
As illustrated in FIG. 16, the second evaluation unit 250 of Modification 3 includes a work determination unit 270 and a work comparison unit 272.
The work determination unit 270 determines the work content of a person based on the skeleton image of the person detected by the skeleton detection unit 222. The work content is determined, for example, by matching the work area in which the skeleton image exists and the shape of the skeleton image.
The work comparison unit 272 compares the skeleton image in the normal work of the work content determined by the work determination unit 270 with the detected skeleton image, and the detected skeleton image can be said to be a normal work. Determine if there is. Whether or not the work is a normal work is determined based on, for example, the time required for each motion and the ratio of unnecessary motions included in the work motion.

FIG. 17 is a flowchart illustrating the work content comparison process (S30).
As shown in FIG. 17, in step 300 (S300), the skeleton detection unit 222 (FIG. 14) detects an image (skeleton image) corresponding to a human skeleton from the images captured by the tablet terminal 4. ..
In step 305 (S305), the work determination unit 270 (FIG. 16) determines the work content of the person based on the skeleton image detected by the skeleton detection unit 222.
In step 310 (S310), the work comparing unit 272 compares the detected skeleton image with the skeleton image in the normal work having the work content determined by the work determining unit 270, and the detected skeleton image is When it is determined whether or not it can be called normal work, and when it is determined that it is normal work, the process proceeds to S315, and when it is determined that it is not normal work Then, the process proceeds to S320.

In step 315 (S315), the second evaluation unit 250 outputs the second evaluation value indicating that the work is normal.
In step 320 (S320), the second evaluation unit 250 outputs a second evaluation value indicating that the work is abnormal.
In step 325 (S325), the second evaluation unit 250 measures the duration of the abnormal work and outputs the duration of the abnormal work as the second evaluation value.
When it is determined that the work is abnormal, the display unit 240 according to the modification 3 arranges and displays the duration of the work in a state where it can be compared with the first evaluation value.

1... Productivity evaluation system 2... Productivity evaluation device 22, 23, 24... Productivity evaluation program 4... Tablet terminal

Claims (9)

A camera installed to shoot the manufacturing site,
A position specifying unit that specifies the position of a person from the image taken by the camera,
A productivity evaluation system including: an evaluation unit configured to generate an evaluation value for evaluating an operating rate of the manufacturing equipment installed at the manufacturing site based on the position of the person specified by the position specifying unit. Further comprising a person extraction unit for extracting a characteristic image of a person from the images taken by the camera,
The position specifying unit specifies the position of the characteristic image of the person extracted by the person extracting unit,
The evaluation unit determines whether or not a characteristic image of a person exists in the area set with reference to at least one of the installation position and the installation direction of the manufacturing facility, and determines the number of people existing in the set area. The productivity evaluation system according to claim 1, wherein the evaluation value is generated based on. The camera captures a plurality of still images that are temporally continuous, or a moving image,
The productivity evaluation system according to claim 2, wherein the evaluation unit calculates the staying time and the number of people existing in the set area as the evaluation value. The productivity evaluation system according to claim 3, further comprising: a display unit that superimposes and displays an evaluation value image indicating an evaluation value generated by the evaluation unit on any image captured by the camera. A voice at the manufacturing site, or a detection unit that detects electric power consumed by the manufacturing facility,
A second evaluation unit that generates a second evaluation value for evaluating the operating rate of the manufacturing facility based on the voice or the power detected by the detection unit;
The production according to claim 3, further comprising: a display unit configured to display the staying time or the number of persons calculated by the evaluation unit and the second evaluation value generated by the second evaluation unit in a comparable manner. Sex evaluation system. A scanner device that scans the business report described by the worker at the manufacturing site,
A description extraction unit that extracts description items from the business report scanned by the scanner device,
The productivity evaluation according to claim 3, further comprising: a display unit that displays, in association with each other, the description items extracted by the description extraction unit and the stay time or the number of people calculated by the evaluation unit based on a time axis. system. From the image taken by the camera installed at the manufacturing site, the position specifying unit that specifies the position of the person,
A productivity evaluation device having an evaluation unit for generating an evaluation value for evaluating the operating rate of the manufacturing equipment installed at the manufacturing site based on the position of the person specified by the position specifying unit. The camera shoots the manufacturing floor,
A computer specifying the position of a person from the images taken by the camera;
A computer generates an evaluation value for evaluating an operating rate of the manufacturing facility installed at the manufacturing site based on the position of the specified person. From the image taken by the camera installed to shoot the manufacturing site, the step of specifying the position of the person,
And a step of generating an evaluation value for evaluating the operating rate of the manufacturing equipment installed at the manufacturing site based on the position of the specified person.

Images