JP2021125183A - Workload analysis apparatus, workload analysis method, and program - Google Patents

Abstract

To provide a more accurate object tracking recovery technology.SOLUTION: A workload analysis apparatus which visualizes load of a worker in a work process includes: a sensor for monitoring a state of the worker; a determination unit which determines whether the worker is working on the basis of data acquired by the sensor; and a display unit which displays workload of the worker on the basis of working time and non-working time of the worker in a predetermined period.SELECTED DRAWING: Figure 2

Description

The present invention relates to a technique for analyzing and visualizing a load for each work process in a factory.

In processing an assembly line consisting of a plurality of elemental works, it is required to appropriately distribute the work so that the load is not concentrated on a specific process. Further, in the cell production method in which one person performs a plurality of processes, it is necessary to avoid delaying the work due to waiting for work in process at another work place.

Conventionally, a method of recommending worker assignment to each work has been proposed based on information such as worker skills, work difficulty, and worker attributes input by the factory manager (Patent Documents). 1, 2). However, it is difficult to judge the skills of workers, and it is difficult to assign them appropriately. In addition, the replacement of workers is also one of the factors that prevent proper allocation.

Japanese Unexamined Patent Publication No. 10-261122 JP-A-2017-68428

If the load for each work process at the actual work site can be visualized, it will help the factory manager to adjust the work allocation and personnel allocation accordingly.

Therefore, an object of the present invention is to provide a technique capable of analyzing and visualizing a worker's load for each work process.

In order to achieve the above object, the present invention adopts the following configuration.

The first aspect of the present invention is
It is a work load analyzer that visualizes the load of workers in the work process.
A sensor for monitoring the worker's condition and
A determination unit that determines whether or not the worker is working based on the data acquired by the sensor, and a determination unit.
A display unit that displays the workload of the worker based on the time during which the worker is working and the time during which the worker is not working in a predetermined period.
Provide a workload analyzer equipped with.

According to this configuration, it is possible to determine whether or not the worker is working at any time, and therefore, it is possible to grasp the time during work and the time during non-work. This makes it possible to quantitatively grasp and visualize the workload of the worker in the work process.

Examples of the sensor include a camera that captures an image of an operator, a motion sensor attached to the operator, and a beacon receiver that detects radio waves transmitted by a beacon possessed by the operator. The camera is, for example, a fisheye camera mounted on the ceiling, and may be one or a plurality of the cameras. By using the camera, it is possible to determine whether or not the work is in progress without imposing a burden on the operator and at low cost. A motion sensor can be used to accurately detect the movement of an operator. Beacon receivers can be used to detect the location of workers.

The determination unit may determine whether or not the worker is working based on the total amount of movement of the specific part of the worker in a predetermined period based on the data acquired by the sensor. good. Specific parts include, for example, wrists, palms, heads, and torso, but are not limited to specific parts. The specific part may be determined according to the work process. Further, the determination unit may make a determination based on the total amount of movement of the plurality of parts.

For example, the determination unit has a posture detection unit that detects the skeleton point of the worker from the image captured by the camera, and obtains the movement amount of the specific part of the worker from the information of the detected skeleton point. You may. In addition, the determination unit may obtain the amount of movement of the specific part of the worker based on the data acquired by the motion sensor.

This configuration is suitable for determining a work in which the movement of a specific part is large during the work.

The determination unit has a human body detection unit that detects the human body of the worker from an image captured by the camera, and the worker is based on the total difference between frames in the human body region in a predetermined frame period. May determine if is in the process of working. Since the magnitude of the difference between frames can be regarded as the magnitude of movement, this method can also be considered as a judgment based on the amount of movement of the operator. That is, the same effect as the above configuration can be obtained. Instead of based on the inter-frame difference of the entire human body region, the determination may be made based on the inter-frame difference in a specific region of the human body region.

Based on the data acquired by the sensor, the determination unit determines that the work has started when the predetermined first action is detected, and ends the work when the predetermined second action is detected. You may judge that. The first action is preferably an action performed at the start of the work process, and the second action is preferably an action performed at the end of the work process. For example, the first action may be an action of acquiring a work, and the second action may be an action of placing a work.

Specifically, the determination unit has a posture detection unit that detects the skeleton point of the worker from the image captured by the camera, and the first action of the worker from the information of the detected skeleton point. Alternatively, the second action may be detected. Further, the determination unit may detect the first action or the second action of the worker based on the data acquired by the motion sensor.

Based on the data acquired by the sensor, the determination unit determines that the work has started when the worker's entry into the predetermined first area is detected, and determines that the work has started and enters the predetermined second area. It may be determined that the work is completed when is detected. The first action is preferably an area in which the worker moves at the start of the work process, and the second action is preferably an area in which the worker moves at the end of the work process. The second region may be set as an region other than the first region.

Specifically, the determination unit has a human body detection unit that detects the human body of the worker from an image captured by the camera, and the first unit of the worker is based on the detected position of the human body. The entry into the region or the second region may be detected. Further, the determination unit may detect the invasion of the worker into the first region or the second region based on the detection result of the beacon receiver.

The determination unit has a trained model that has been machine-learned to output whether or not the worker is working, and inputs the data acquired by the sensor into the trained model. It may be determined whether or not the worker is working. The sensor is arbitrary such as a camera, a motion sensor, and a beacon receiver as described above. Further, a combination of data obtained from a plurality of types of sensors may be input to the trained model.

In this embodiment, the determination unit may make a determination for each of the plurality of work processes, and the display unit may display the workload for each work process. This display may include, for example, a graph showing whether or not the worker was working at each time of a predetermined period for each work process, and for each work process from the start to the end of the work. It may include information representing the average time required. According to this configuration, it is possible to compare the workload for each work process.

The present invention may be regarded as a workload analysis device having at least a part of the above means, or may be regarded as a workload evaluation device, a workload monitoring device, or the like. Further, the present invention can be regarded as a method including at least a part of the above processing, a program for realizing such a method, or a recording medium in which the program is recorded non-temporarily. It should be noted that each of the above means and treatments can be combined with each other as much as possible to form the present invention.

According to the present invention, the load on the worker for each work process can be analyzed and visualized.

The figure which shows the application example of the workload analyzer which concerns on this invention. The figure which shows the structure of the workload analyzer which concerns on 1st Embodiment. Flowchart of processing performed by the workload analyzer. An example of a graph showing the workload for each work process. The figure which shows the structure of the workload analyzer which concerns on 2nd Embodiment. The figure which shows the structure of the workload analyzer which concerns on 3rd Embodiment.

<Application example>
One of the application examples of the present invention will be described with reference to FIG. FIG. 1 shows an example in which the workload analyzer 10 according to the embodiment of the present invention is applied to load analysis of a work process in a production line such as a factory.

In the example of FIG. 1, a camera 20 is installed on the ceiling of the work place, and an image is taken of a production line including a plurality of workers P1 and P2. The image captured by the camera 20 is captured by the workload analyzer 10. The workload analyzer 10 analyzes the image and determines whether or not the worker is working. By continuously determining whether or not work is in progress, the workload for each process can be obtained. The calculated workload for each process is displayed on the display device 21.

According to such a workload analyzer 10, it is possible to easily confirm which work process the load is applied to. Therefore, the manager can determine the work allocation and staffing of each process in consideration of the actual workload.

The object to which the workload can be analyzed by the present invention is not limited to the workplace of the line production method, but can be applied to the workplace of the cell production method. Further, the present invention is not limited to the workplace where a product is manufactured, and can be applied to a scene in which an arbitrary work is performed.

<First Embodiment>
A specific configuration example of a workload analysis system to which the workload analysis device 10 according to the embodiment of the present invention is applied will be described with reference to FIG.

(composition)
The workload analyzer 10 includes one or more processors, a main storage device, an auxiliary storage device, a communication device, an input device, and an output device as hardware components thereof, and the processor executes a computer program to: Execute various processes of. Note that some or all of the processing may be executed by a dedicated hardware circuit.

The workload analysis device 10 of the present embodiment has an image acquisition unit 11, a work determination unit 12, a graph generation / display unit 16, and a storage unit 15 as its functional units. The image acquisition unit 11 acquires an image from the camera 20. The work determination unit 12 determines whether or not the worker is working based on the camera image, and stores the determination result in the storage unit 15. The work determination unit 12 includes a human body detection unit 13 and a posture detection unit 14 as sub-function units. The human body detection unit 13 detects a region in which the human body is captured from the camera image. The posture detection unit 14 detects the skeleton point of the human body (worker) from the camera image. The work determination unit 12 determines whether or not the worker is working based on at least one of the human body detection result of the human body detection unit 13 and the posture (skeleton point) detection result of the posture detection unit 14. In the embodiment, the start time and the end time of the work are stored in the storage unit 15. The graph generation / display unit 16 generates a graph showing the workload of each work process based on the work start time and work end time stored in the storage unit 15, and displays the graph on the display unit 16. The details of the workload analyzer 10 will be described below together with the flowchart.

The camera 20 is an image pickup device having an optical system including a fisheye lens and an image pickup device (an image sensor such as a CCD or CMOS). As shown in FIG. 1, for example, the camera 20 may be installed on the ceiling of a work place or the like with the optical axis facing vertically downward, and may take an image in all directions (360 degrees). Therefore, the state of the worker can be monitored by the camera 20. The camera 20 is connected to the workload analyzer 10 by wire (USB cable, LAN cable, etc.) or wirelessly (WiFi, etc.), and the image data captured by the camera 20 is taken into the workload analyzer 10. The image data may be either a monochrome image or a color image, and the resolution, frame rate, and format of the image data are arbitrary. In this embodiment, it is assumed that a color (RGB) image captured at 10 fps (10 images per second) is used.

The display device 21 displays the image data output from the workload analysis device 10. The display device 21 may be any kind of display device such as a liquid crystal display, an organic EL display, a CRT display, and a projector.

(process)
FIG. 3 is an overall flowchart of the processing performed by the workload analyzer 10.

In step S11, the image acquisition unit 11 acquires an image (camera image) captured by the camera 20. The image acquisition unit 11 may acquire an image in real time at the same time as the image is captured by the camera 20, or the image stored in the storage device by the camera 20 may be acquired.

The following steps S12 to S14 may be performed for each frame of the camera image, or may be performed at predetermined frame intervals. Further, when a plurality of workers who perform a plurality of work processes are transferred to the camera image, processing is performed for each of the plurality of work processes (workers). Further, when one worker performs a plurality of work processes, each of the plurality of work processes is also processed for one person's work.

In step S12, the human body detection unit 13 detects an estimated region (bounding box) in which the human body exists from the camera image. Human body detection may be performed using any existing algorithm. For example, human body detection can be performed using a classifier learned by statistical machine learning based on HoG (Histogram of Gradient) features or Haar-like features. Further, for example, the human body can be detected by using a discriminator learned by a neural network such as deep learning. Further, a classifier combining a plurality of weak classifiers learned by ensemble learning such as boosting and bagging may be used.

In step S13, the posture detection unit 14 detects a skeleton point (also called a key point) of the human body from the camera image. The posture detection unit 14 detects the positions of dozens to hundreds of skeletal points including the crown, neck, shoulders, elbows, wrists, hip joints, knees, and ankles. It is preferable to use a deep learning-based estimation model for posture detection. The posture detection process may be performed on the entire camera image, or may be performed only on the human body region detected by the human body detection.

In step S14, the work determination unit 12 determines whether or not the worker shown in the camera image is working. There are several conceivable methods for determining the working state of the worker from the camera image.

[First method: Judgment based on the amount of movement of a specific part]
The first method determines whether or not the worker is working based on the total amount of movement of the specific part of the worker within the latest predetermined period.

ここで、p_r(i)は時刻（フレーム）ｉにおける右手の位置（座標）であり、p_l(i)は時刻（フレーム）ｉにおける右手の位置（座標）である。 For example, the determination can be made based on the total amount of movement between the frames of the right hand (right wrist) and the left hand (left wrist) in the last 5 seconds (50 frames). In this case, the total movement amount is calculated by the following formula. If the total movement amount is larger than the threshold value, the work determination unit 12 determines that the work is in progress, and if it is smaller than the threshold value, the work determination unit 12 determines that the work is not in progress.

Here, p _r (i) is the position (coordinates) of the right hand at the time (frame) i, and p _l (i) is the position (coordinates) of the right hand at the time (frame) i.

Here, the total amount of movement of the right hand and the left hand is focused on, but information such as the upper body, lower body, and head may be used. Further, the amount of movement of three or more parts may be used, or the amount of movement of only one part may be used.

[Second method: Judgment based on inter-frame difference]
The second method determines whether or not the worker is working based on the sum of the inter-frame differences of the human body region in the most recent predetermined period.

The work determination unit 12 calculates the total of the absolute value differences (inter-frame differences) of the absolute value differences of each pixel value in the area after performing the process of matching the sizes of the human body regions between the frames. The square of the difference may be used instead of the absolute value difference. The work determination unit 12 determines that the work is in progress if the total of the inter-frame differences in the human body region in the predetermined frame period is larger than the threshold value, and determines that the work is not in progress if it is smaller than the threshold value.

[Third method: Judgment based on detection of specific motion]
The third method determines whether or not the worker is working by detecting that the worker has performed a specific action (behavior).

The work determination unit 12 determines that the work has started when a characteristic action (first action) to be performed at the start of the work process is detected, and the work determination unit 12 determines that the work has started and performs a characteristic action (second action) to be performed at the end of the work process. Is detected, it is determined that the work has started. Examples of characteristic movements at the start of work include movements such as picking up a work handed over from an upstream process and picking up a predetermined tool. Further, as an example of a characteristic operation at the end of work, there are operations such as placing a work in a predetermined place and placing a predetermined tool. These are examples, and the working state may be determined based on other operations. For example, when an action of moving a part such as an arm or a leg in a predetermined pattern is detected, it may be determined that the work has started or the work has ended.

The motion may be detected based on the skeleton point information detected by the posture detection unit 14. In addition, the movement of picking up or placing a work or tool can be detected by using an object detection technique for detecting these objects.

[Fourth method: Judgment based on intrusion into a specific area]
The fourth method determines whether or not the worker is working based on the position of the worker.

The work determination unit 12 determines that the work has started when the worker has invaded a predetermined area (first area), and has determined that the work has been completed when the worker has invaded another predetermined area (second area). do. For example, in a cell production method in which one worker performs a plurality of processes, if the work is performed in a different place for each work process, the place (area) where the target work process is performed corresponds to the first area, which corresponds to the first area. Areas other than the second area correspond to the second area. Further, in one work process, when a work is acquired from a certain place at the start of work and a work is placed in another place at the end of work, the place where the work is acquired corresponds to the first area and the place where the work is placed. Corresponds to the second region.

The position of the worker can be acquired from the detection result by the human body detection unit 13. The three-dimensional location where the operator is located can be obtained from the shooting conditions of the camera 20 and the position and size of the human body region in the image.

[Fifth method: Judgment using AI model]
The fifth method is whether or not the worker is working by using a trained model (AI model) that has been machine-learned in advance to receive information about the worker as input and output the work state of the worker. Is determined.

The input data to the training model includes at least one of the skeleton point information and the frame image in the most recent predetermined period. The learning model can be generated by supervised learning using teacher data to which a correct answer flag indicating whether or not the work is being performed is added.

The work determination unit 12 determines the work state of the worker by using at least one of the methods described above. When a plurality of determination methods are used, it may be determined that the work is in progress when it is determined that the work is in progress by a predetermined number or more of the methods. The predetermined number may be 1 or all. Further, when a plurality of determination methods are used, the working state may be determined based on a value obtained by weighting and averaging each determination result according to the reliability. Depending on the determination method to be adopted, the work determination unit 12 may not have either or both of the human body detection unit 13 and the posture detection unit 14.

In step S15, the work determination unit 12 stores the work start time and the work end time in the storage unit 15. The work determination unit 12 stores the work state (whether working or non-working) by the worker in the immediately preceding frame. When the work state changes from non-working to working in the current frame, the work determination unit 12 stores the time of the current frame as the work start time in the storage unit 15. Further, when the work determination unit 12 changes the work state from working to non-work, the time of the current frame is stored in the storage unit 15 as the work end time. The work start time and work end time are stored in association with the work process.

Here, the work start time and the work end time are memorized, but the work state of the worker for each frame may be memorized.

In step S16, the graph generation / display unit 16 generates a graph representing the workload based on the time during which the worker is working and the time during which the worker is not working for each work process. Specifically, the graph generation / display unit 16 obtains the working time and the non-working time based on the work start time and the work end time for each work process stored in the storage unit 15, and works. Generate a graph showing the workload for each process.

In step S17, the graph generation / display unit 16 displays the generated graph on the display device 21.

4A and 4B show examples of graphs generated and displayed by the graph generation / display unit 16.

FIG. 4A is an example of a graph 401 showing whether or not the worker was working at each time of a predetermined period for each work process. Since the work determination unit 12 can know the work state at each time from the work start time and the work end time stored in the storage unit 15, the graph 401 can be generated. From such a graph, the load of each work process can be easily grasped. In the example of FIG. 4A, it can be easily grasped that the load of the process B and the process D is high, the load of the process C is low, and the load of the process A is standard.

FIG. 4B is an example of a graph 402 displaying the average time required from the start of the work to the end of the work for each work process. The work determination unit 12 can generate the graph 402 by obtaining the time between the work start time and the work end time stored in the storage unit 15 and averaging them. Graph 402 also includes a display showing standard working hours. From such a graph, the load of each work process can be easily grasped. In the example of FIG. 4B, it can be easily understood that the load in step 3 is high, the load in steps 2 and 5 is low, and the load in steps 1 and 4 is standard.

The workload for each work process does not necessarily have to be indicated by a graphical display, and may be indicated by characters (text). The workload can be grasped even in the text display. In addition, the workload can be grasped more easily by changing the color and size of the characters and adding character decoration according to the workload.

(Advantageous effect of this embodiment)
According to this embodiment, since the workload for each work process is visualized, the factory manager can adjust the work allocation and the personnel allocation based on the actual workload.

Further, in the present embodiment, the working state of the worker is determined based on the information from the camera 20 provided on the ceiling, and there is an advantage that the worker is not burdened.

<Second Embodiment>
In the first embodiment, the state of the worker is monitored by the camera 20 installed in the work place. In the present embodiment, the state of the worker is monitored by using the motion sensor attached to the worker.

FIG. 5 is a diagram showing a configuration of a workload analysis system to which the workload analysis device 50 according to the present embodiment is applied. Since the workload analyzer 50 is basically the same as the workload analyzer 10 of the first embodiment, different parts will be mainly described.

The workload analysis device 50 has a motion data acquisition unit 17 that acquires motion data from a motion sensor 22 attached to the operator instead of the image acquisition unit 11. The motion sensor 22 is attached to one or a plurality of parts of the worker, and transmits the motion information of the part to the motion data acquisition unit 17 by wireless communication. The motion sensor 22 is typically a 9-axis sensor that combines a 3-axis acceleration sensor, a 3-axis angular velocity sensor (gyro sensor), and a 3-axis geomagnetic sensor, but may be composed of only a 3-axis acceleration acceleration sensor. ..

In the present embodiment, the movement of a specific part of the operator can be grasped by the motion data obtained from the motion sensor 22. For example, if the motion sensor 22 is attached to the right hand and the left hand of the worker, respectively, the workload analyzer 50 can acquire the movements of the right hand and the left hand of the worker. Therefore, in the present embodiment, the work determination unit 12 calculates the total amount of movement of the specific part in a predetermined period based on the movement of the specific part obtained from the motion data, and the operator works based on the total amount of movement. It can be determined whether or not it is inside (first method). Alternatively, the work determination unit 12 may detect a specific motion based on the movement of the specific portion obtained from the motion data, and determine whether or not the worker is working from the detection result (third). the method of). Further, the work determination unit 12 may input motion data into the learning model and determine whether or not the worker is working (fifth method).

According to the present embodiment, in the first embodiment, the work state determination based on the movement amount of the specific part, the work state determination based on the detection of the specific motion, and the work state determination using the AI model can be performed.

Although the camera is not adopted in the above description, the camera may be further used to determine the working state by using both the camera image and the motion data. Further, instead of using the motion sensor, the motion of the operator may be acquired by using the motion capture technique. That is, the movement of the portion to which the marker is attached may be detected by attaching the marker to the operator and detecting the marker with a plurality of cameras.

<Third Embodiment>
In the present embodiment, the state (position) of the worker is monitored by detecting the radio wave transmitted from the beacon transmitter attached to the worker.

FIG. 6 is a diagram showing a configuration of a workload analysis system to which the workload analysis device 60 according to the present embodiment is applied. Since the workload analyzer 60 is basically the same as the workload analyzer 10 of the first embodiment, different parts will be mainly described.

In addition to the configuration of the workload analyzer 10, the workload analyzer 60 further includes a beacon receiver 18 that detects radio waves transmitted by a beacon transmitter owned by the operator. The beacon transmitter 23 emits a radio wave including identification information, and the beacon receiver 18 can grasp which beacon (worker) is located where by receiving this radio wave, that is, a worker to a predetermined area. Can detect the invasion of. Therefore, the work determination unit 12 detects the entry of the worker into the first region or the second region based on the detection result of the radio wave by the beacon receiver 18, and determines the working state of the worker based on the detection result. It can be determined (fourth method).

The position of the worker may be more accurately detected by using a beacon than by detecting it based on a camera image, and this embodiment is effective in such a case.

Although the camera 20 is used in the above description, the camera 20, the human body detection unit 13, and the posture detection unit 14 can be omitted when the working state is determined based only on the approach to the specific area. be. Further, this embodiment may be implemented in combination with the second embodiment. Further, in addition to the beacon, the position of the operator may be detected by a motion sensor (infrared sensor, ultrasonic sensor, or a combination thereof).

<Others>
The above-described embodiment merely exemplifies a configuration example of the present invention. The present invention is not limited to the above-mentioned specific form, and various modifications can be made within the scope of its technical idea.

<Additional notes>
1. 1. It is a work load analyzer (10, 50, 60) that visualizes the load of workers in the work process.
Sensors (20, 22, 23) for monitoring the worker's condition and
A determination unit (12) that determines whether or not the worker is working based on the data acquired by the sensor, and a determination unit (12).
A display unit (16) that displays the workload of the worker based on the working time and the non-working time of the worker in a predetermined period.
A workload analyzer equipped with.

2. It is a work load analysis method that visualizes the load of workers in a work process performed by a computer (10, 50, 60).
A determination step (S14) for determining whether or not the worker is working based on the data acquired by the sensors (20, 22, 23) that monitor the state of the worker, and a determination step (S14).
A display step for displaying the workload of the worker based on the working time and the non-working time of the worker in a predetermined period (S16, S17).
Workload analysis methods including.

10: Workload analyzer 11: Image acquisition unit 12: Work judgment unit 13: Human body detection unit 14: Posture detection unit 15: Storage unit 16: Graph generation / display unit 17: Motion data acquisition unit 18: Beacon receiver 20: Camera 21: Display device 22: Motion sensor 23: Beacon transmitter

Claims (17)

It is a workload analyzer that visualizes the load of workers in the work process.
A sensor for monitoring the worker's condition and
A determination unit that determines whether or not the worker is working based on the data acquired by the sensor, and a determination unit.
A display unit that displays the workload of the worker based on the time during which the worker is working and the time during which the worker is not working in a predetermined period.
A workload analyzer equipped with. The determination unit determines whether or not the worker is working based on the total amount of movement of the specific part of the worker in a predetermined period based on the data acquired by the sensor.
The workload analyzer according to claim 1. The sensor is a camera that captures the worker.
The determination unit has a posture detection unit that detects the skeleton point of the worker from the image captured by the camera, and obtains the movement amount of the specific part of the worker from the information of the detected skeleton point.
The workload analyzer according to claim 2. The sensor is a motion sensor attached to the worker.
The determination unit obtains the amount of movement of the specific part of the worker based on the data acquired by the motion sensor.
The workload analyzer according to claim 2. The sensor is a camera that captures the worker.
The determination unit has a human body detection unit that detects the human body of the worker from an image captured by the camera, and the worker is based on the total difference between frames in the human body region in a predetermined frame period. To determine if is working,
The workload analyzer according to any one of claims 1 to 4. Based on the data acquired by the sensor, the determination unit determines that the work has started when the first action is detected, and determines that the work has ended when the second action is detected.
The workload analyzer according to any one of claims 1 to 5. The sensor is a camera that captures the worker.
The determination unit has a posture detection unit that detects the skeleton point of the worker from the image captured by the camera, and the first action or the second action of the worker from the information of the detected skeleton point. To detect,
The workload analyzer according to claim 6. The sensor is a motion sensor attached to the worker.
The determination unit detects the first action or the second action of the worker based on the data acquired by the motion sensor.
The workload analyzer according to claim 6. Based on the data acquired by the sensor, the determination unit determines that the work has started when the worker's entry into the first region is detected, and the entry into the second region is detected. If you decide that the work is done,
The workload analyzer according to any one of claims 1 to 8. The sensor is a camera that captures the worker.
The determination unit has a human body detection unit that detects the human body of the worker from an image captured by the camera, and the first region or the second region of the worker is based on the detected position of the human body. Detecting entry into,
The workload analyzer according to claim 9. The sensor is a beacon receiver that detects radio waves transmitted by a beacon possessed by the worker.
The determination unit detects the worker's entry into the first region or the second region based on the detection result of the beacon receiver.
The workload analyzer according to claim 9. The determination unit has a trained model that has been machine-learned to output whether or not the worker is working, and inputs the data acquired by the sensor into the trained model. Determine if the worker is working,
The workload analyzer according to any one of claims 1 to 11. The determination unit makes a determination for each of the plurality of work processes.
The display unit displays the workload for each work process.
The workload analyzer according to any one of claims 1 to 12. The display unit displays a graph showing whether or not the worker was working at each time of a predetermined period for each work process.
The workload analyzer according to claim 13. The display unit displays information representing the average time required from the start of the work to the end of the work for each work process.
The workload analyzer according to claim 13 or 14. It is a work load analysis method that visualizes the load of workers in the work process performed by a computer.
A determination step for determining whether or not the worker is working based on the data acquired by the sensor that monitors the state of the worker, and a determination step.
A display step for displaying the workload of the worker based on the working time and the non-working time of the worker in a predetermined period.
Workload analysis methods including. A program for causing a computer to perform each step of the method according to claim 16.

Images