Work content analysis device, work content analysis method, program, and sensor

Abstract

To analyze work contents of workers of a plant or the like without imposing excessive labor costs and without requiring remarkable modification of facilities.SOLUTION: According to an embodiment, a work content analysis device for analyzing work contents of a worker on the basis of a state of the worker, includes: a first database for associating state information representing a state of the worker with time information and identification information of the worker to store the state information; an estimation part for estimating work contents performed by the worker from at least two pieces of state information associated at the same time from among pieces of state information stored in the first database; an identification part for identifying a work time spent for the estimated work contents from the state information stored in the first database and the time information associated with the state information; and an analysis part for analyzing the work contents on the basis of the estimated work contents and the identified work time.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] An embodiment of the present invention relates to a work content analysis device, a work content analysis method, a program, and a sensor for the work content analysis device, for analyzing work content of a worker in a factory or the like based on the state of the worker. [Background technology]

[0002] Typically, to improve work efficiency, the time required for actual work is measured, the work procedures are analyzed, and based on the results of this measurement and analysis, the current state of work is visualized, the procedures are standardized, and each worker follows the standardized procedures.

[0003] However, the measurement of the work time required for such a conventional analysis of the work procedure is performed by a measurer using a stopwatch, which requires additional labor costs.

[0004] IoT can also be used to measure work time without incurring extra labor costs. In other words, the idea is to use IoT to replace human-based work.

[0005] However, heavy-duty IoT requires modifications to the factory equipment itself, making it difficult to retrofit equipment in an already operating factory.

[0006] Thus, even if one were to conduct a work content analysis, collecting the data for it would require extra labor costs, and conversely, if one were to try to reduce labor costs, it would require modifying the facilities, making it no easy task.

[0007] For this reason, there is not enough capacity to implement measures to improve work efficiency at all of the factories involved, and the current situation is that each factory is left to decide on its own measures to improve work efficiency. [Prior art documents] [Patent documents]

[0008] [Patent Document 1] Patent No. 4218669 [Patent Document 2] Patent No. 5198981 [Patent Document 3] Patent No. 5107206 [Patent Document 4] Patent No. 6520029 [Patent Document 5] Patent No. 6385613 [Patent Document 6] Patent No. 5159263 Summary of the Invention [Problem to be solved by the invention]

[0009] The problem to be solved by the present invention is to provide a work content analysis device, a work content analysis method, and a program for analyzing the work content of workers in a factory or the like, without incurring extra labor costs or requiring major modifications to the facilities, and also to provide a sensor to be used for the work content analysis device. [Means for solving the problem]

[0010] A task analysis device according to an embodiment is a task analysis device that analyzes task contents of a worker based on the state of the worker, and includes a first database that stores state information representing the state of one or more workers in association with time information and identification information of the worker, an estimation unit that estimates task contents performed by the worker from at least two pieces of state information associated with the same time among the state information stored in the first database, an identification unit that identifies the task time spent on the estimated task content from the state information stored in the first database and the time information associated with the state information, and an analysis unit that analyzes the task content based on the estimated task content and the identified task time. The state information includes at least one of the position information of the worker, the movement information indicating whether the worker's hands are moving, and the information indicating the operating state of equipment around the worker. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is a block diagram showing an example of an electronic circuit configuration of a work content analysis device to which a work content analysis method according to a first embodiment is applied. [Figure 2A] FIG. 2A is a conceptual diagram showing an example of a connection relationship (including a camera) between the work content analysis device and other devices. [Figure 2B] FIG. 2B is a conceptual diagram showing an example of a connection relationship (including facilities) between the work content analysis device and other devices. [Figure 2C] FIG. 2C is a block diagram showing an example of an electronic circuit configuration of the sensor. [Figure 3A] FIG. 3A is an example showing, in time series, status information (sensing information indicating the operating status of equipment) conventionally stored in a status information database. [Figure 3B] FIG. 3B is another example showing, in time series, state information (sensing information indicating the operating state of the facility) stored in the state information database. [Figure 4A] FIG. 4A is a diagram showing an example of a time-series change in sensing information indicating an area number as position information. [Figure 4B]FIG. 4B is a diagram showing an example of state information determined by performing arithmetic processing on the sensing information shown in FIG. 4A. [Diagram 5] FIG. 5 is a flowchart showing an example of the operation of a work content analysis device to which the work content analysis method of the first embodiment is applied. [Figure 6] FIG. 6 is an example of a graphic display showing the time required for each task. [Figure 7] FIG. 7 is a flowchart for explaining the flow of business improvement. [Figure 8] FIG. 8 is a flowchart showing an example of the operation of a task content analysis device to which the task content analysis method of the second embodiment is applied. [Figure 9] FIG. 9 is an example of a graphic display showing the frequency and duration of visits to each area of ​​a factory by all workers in a group. [Figure 10] Figure 10 is an example of a graphic display showing the actual work hours broken down by work type. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0013] (First embodiment) A task content analysis device to which a task content analysis method according to a first embodiment is applied will be described.

[0014] FIG. 1 is a block diagram showing an example of an electronic circuit configuration of a work content analysis device to which a work content analysis method according to a first embodiment is applied.

[0015] FIG. 2A is a conceptual diagram showing an example of a connection relationship (including a camera) between the work content analysis device and other devices.

[0016] The work content analysis device 10 is a device that analyzes the work content performed by workers in a factory or the like.

[0017] As shown in FIG. 1, the electronic circuitry of the work content analysis device 10 includes a CPU 12, a recording medium reading unit 14, a communication unit 15, a display unit 16 (e.g., a display), a memory 20, and a storage device 30, all connected to each other by a bus 11.

[0018] The memory 20 stores a processing module 21, an estimation module 22, a specification module 23, a task time comparison module 24, an analysis module 25, a display control module 26, and a judgment module 27 as programs for implementing the task content analysis device 10.

[0019] These program modules 21 to 27 may be pre-stored in the memory 20, or may be read and stored in the memory 20 from an external recording medium 13 such as a memory card via the recording medium reading unit 14. These program modules 21 to 27 are designed not to be rewritable.

[0020] In addition to such an area that cannot be rewritten by the user, a writable data area 29 is secured in the memory 20 as an area for storing rewritable data.

[0021] The CPU 12 is an example of one or more processors capable of executing each of the program modules 21-27, and controls the operation of each part of the circuit in accordance with each of the program modules 21-27.

[0022] The storage device 30 includes a worker database 31 , a status information database 32 , and a reference time database 33 .

[0023] The worker database 31 stores, for example, the worker ID (for example, employee number) of each worker 100 in a factory.

[0024] The status information database 32 stores sensing information from each sensor 110 attached to each worker 100 as status information representing the status of each worker 100 in association with time information, a worker ID, and a sensor ID.

[0025] The status information database 32 can also store sensing information from sensors 110, such as cameras, installed near or around the workers 100 as status information for each worker 100, in association with time information, worker ID, and sensor ID. In this case, however, the sensor 110 is stored in association with the worker ID of a worker 100 who is not attached to the worker 100 but is near the sensor 110, such as a worker 100 who is within a predetermined distance.

[0026] The status information may include position information of worker 100, movement information indicating whether worker 100's hands are moving, information indicating the operating status of equipment around worker 100, movement information indicating whether worker 100 is moving, vital sign information of worker 100, audio information of worker 100, ambient sound information of worker 100, and video information of the surroundings of worker 100.

[0027] Furthermore, the status information database 32 can store not only sensing information (hereinafter also referred to as "first status information") acquired from the sensor 110 (including the camera 115) as status information, but also status information (hereinafter also referred to as "second status information") obtained by performing mathematical calculation processing on the sensing information (first status information) in association with time information, a worker ID, and a sensor ID as status information representing the status of each worker 100. The status information is not limited to sensing information (first status information) from the sensor 110, but can also include image information from the camera 115 and, for example, a signal from a patrol lamp.

[0028] The status information can be obtained not only as sensing information transmitted from each sensor 110 attached to each worker 100 or a camera 115 installed at a fixed point and controlled by a microcontroller 120, but also based on sensing information, such as the second status information obtained from the first status information described above.

[0029] Each worker 100 can be fitted with a plurality of types of sensors 110, not just one type.

[0030] These sensors, including sensor 110 and camera 115, are IoT devices for visualizing the movements of the worker, and specific types include, but are not limited to, a position sensor that measures the position of worker 100 via GPS function, WiFi function, Bluetooth (registered trademark) function, or beacon 125, an acceleration sensor that detects the hand movements of worker 100, a vital sensor that measures vital information such as the pulse of worker 100, a microphone that detects the voice of worker 100, a sound sensor that detects sounds around worker 100, and an imaging sensor such as a camera that is attached to the worker itself and captures images in the line of sight of worker 100.

[0031] These sensors are attached to the front of the body of the worker 100, the left and right sides of the waist, the arms, wrists, neck, shoulders, ears, head, etc. so as to easily receive radio waves from the beacon 125. In particular, the sensor 110 for acquiring movement information, such as the acceleration sensor described above, is attached to the dominant arm or hand of the worker 100.

[0032] A wearable terminal including a smartphone can also be used as the position sensor. In this case, the smartphone is placed in a waist holder and secured with a belt, or the hook is hooked onto a belt loop. If the smartphone is worn on the back, it becomes difficult for the communication unit 15 to receive signals from the smartphone, so it is preferable to wear the smartphone on the front of the worker 100 or on the left or right side of the waist. The smartphone may also be carried in a trouser pocket or breast pocket of the worker 100.

[0033] A device on the wristband side can be used as the vital sensor. In this case, since wearing the device on both arms or hands of the worker 100 may cause a hindrance to the work, it is preferable to wear the vital sensor on the arm or wrist of the dominant arm or hand of the worker 100 so that the movement and work content of the worker 100 can be captured in detail with a minimum configuration.

[0034] A headset type device can be used as the sound sensor. In this case, the device can be worn around the neck or on the shoulder of the worker 100.

[0035] An earphone-type device can be used as the sound sensor. In this case, the sound sensor can be attached to the ear of the worker 100.

[0036] A smart glass type device can be used as the image sensor. In this case, the image sensor can be attached to the head of the worker 100.

[0037] In addition, although not attached to the worker 100, for example, a camera 115 installed at a fixed point is also included in the sensors, but is not limited to this.

[0038] FIG. 2B is a conceptual diagram showing an example of a connection relationship (including facilities) between the work content analysis device and other devices.

[0039] A method of acquiring operation information of equipment around the worker 100, out of the state information, will be described with reference to FIG. 2B.

[0040] The connection relationship shown in Fig. 2B differs from that shown in Fig. 2A in that equipment 116 is provided instead of camera 115, but the other configurations are the same. Therefore, the following will describe the differences from Fig. 2A.

[0041] The equipment 116 is controlled by the microcontroller 120 to transmit operation information of the equipment 116 to the communication unit 15. In addition, sensors (not shown) for acquiring operation information may be installed in the equipment 116, and the operation information acquired from these sensors may be transmitted to the communication unit 15.

[0042] In this way, the work content analysis device 10 can acquire operation information of the equipment 116 around the worker 100, which is one piece of state information.

[0043] FIG. 2C is a block diagram showing an example of an electronic circuit configuration of the sensor.

[0044] The block diagram illustrated in FIG. 2C is a generalized example of one of the many types of sensors 110 discussed above.

[0045] The sensor 110 generally includes a CPU 1102, a sensing unit 1103, a transmitting unit 1104, a receiving unit 1105, a memory 1110, and a storage device 1120, all connected together by a bus 1101.

[0046] The CPU 12 is a processor that controls the operation of each of the above-mentioned circuits connected to each other via the bus 1101 .

[0047] The sensing unit 1103 is a part that acquires sensing information of the worker 100. For example, the sensing unit 1103 corresponds to a detection unit that detects acceleration when the sensor 110 is an acceleration sensor, a detection unit that detects vital values ​​such as pulse rate when the sensor 110 is a vital sensor, a sound collection unit that detects sound when the sensor 110 is a sound sensor, an imaging element when the sensor 110 is a camera, or a GPS function, WiFi function, or Bluetooth function when the sensor 110 is a position sensor.

[0048] The sensor 110 can also function as a position sensor that measures the position of the worker 100 via the beacon 125. In that case, the receiving unit 1105 receives a signal from the beacon 125 as sensing information. Note that the beacon 125 can change the intensity of the radio wave it transmits.

[0049] When the sensing unit 1103 and the receiving unit 1105 acquire the sensing information in this manner, they output the sensing information to the storage device 1120. The storage device 1120 includes a sensing information database 1111 for storing the sensing information. 1 21 and stores the sensing information output from the sensing unit 1103 and the receiving unit 1105.

[0050] The storage device 1120 is, for example, a solid state drive (SSD) or a hard disk drive (HDD).

[0051] The memory 1110 stores a communication determination module 1111 as a program included in the sensor 110 .

[0052] The communication determination module 1111 determines whether the sensor 110 is capable of communicating with the activity content analysis device 10 or not.

[0053] If the communication determination module 1111 does not determine that communication between the sensor 110 and the work content analysis device 10 is possible (i.e., if the communication determination module 1111 determines that communication between the sensor 110 and the work content analysis device 10 is not possible), the sensing information database 1 1 The sensing information continues to be stored in 21 .

[0054] When the communication determination module 1111 determines that communication between the sensor 110 and the activity content analysis device 10 is possible, the transmission unit 1104 transmits the sensing information to the sensing information database 1104 regardless of whether new sensing information has been acquired. 1 The sensing information stored in the sensor 21 is transmitted to the work content analysis device 10 via the communication network 70 at predetermined intervals (for example, every 5 seconds or every 10 seconds).

[0055] Each sensor 110 is assigned a sensor ID, which is identification information for the sensor, and is associated with the worker ID of the worker 100 to which the sensor is attached. When transmitting sensing information as status information to the work content analysis device 10, the transmission unit 1104 transmits the sensing information together with the sensor ID and the worker ID.

[0056] In particular, when the sensor 110 is a camera 115, the transmission unit 1104 transmits the image information, which is the sensing information, to the work content analysis device 10 together with the sensor ID.

[0057] 1, the communication unit 15 of the work content analysis device 10 is connected to the communication network 70, receives the sensor ID, worker ID, and sensing information transmitted from each sensor 110 via the communication network 70, and sends the received sensor ID, worker ID, and sensing information to the processing module 21. Similarly, it receives the sensor ID and image information, which is the sensing information, transmitted from the camera 115, and sends the received sensor ID and image information, which is the sensing information, to the processing module 21.

[0058] The communication unit 15 can also output data necessary for graph display to the external terminal 130 via the communication network 70, as shown in Figures 2A and 2B, in order to display graphs, etc., displayed on the display unit 16 by the display control module 26 from the external terminal 130, as described below.

[0059] In this way, the sensing information transmitted as status information from the sensor 110 includes position information of the worker 100, movement information related to the hand movement of the worker 100, movement information related to the movement of the worker 100, vital information of the worker 100, voice information of the worker 100, ambient sound information of the worker 100, video information in the line of sight of the worker 100, and video information of the worker 100. Note that, as will be described later, the processing module 21 can also obtain the movement information and position information of the worker 100 from the video information from the camera 115.

[0060] The state information database 32 is a database for storing state information, and as the time information, time information measured by an internal clock (not shown) of the work content analysis device 10 can be used. Also, time information synchronized with a clock (timing means) of an external system to which the work content analysis device 10 is connected can be used.

[0061] In addition, if synchronization with the timing means of the connected system is not possible, in order to prevent time discrepancies from occurring between sensors, the time of the work content analysis device 10 itself can be used as the reference time, and each sensor 110 can use the time synchronized with the work content analysis device 10.

[0062] The reference time database 33 stores a reference time determined in advance for each task. The reference time is not limited to, but may be, for example, a target time determined in advance for each task, or the time required by a highly skilled person to complete the task.

[0063] The storage device 30 that stores these databases 31 to 33 is also, like the storage device 1120, formed of, for example, an SSD (Solid State Drive) or an HDD (Hard Disk Drive).

[0064] The processing module 21 receives the sensing information sent from the communication unit 15 together with the sensor ID and, if present, the worker ID, and determines whether or not processing is required for this sensing information. For sensing information that requires processing, the processing module 21 performs the necessary processing, acquires status information, and sends the status information together with the sensor ID and worker ID to the status information database 32. On the other hand, for sensing information that does not require processing, the processing module 21 does not process the sensing information, and sends the sensing information as status information together with the sensor ID and worker ID to the status information database 32.

[0065] An example of sensing information that requires processing is video information from the camera 115. The processing of this video information involves, for example, using AI technology to identify the worker 100 captured in the video information, identify the worker ID of the worker 100, detect movement information and position information of the identified worker 100, and use the detected movement information and position information as status information. In this way, when processing video information, the processing module 21 functions in particular as a video processing module. In addition, there may be cases where sensing information from an acceleration sensor is processed or altered using, for example, AI technology. It should be noted that sensing information that requires processing is not limited to these.

[0066] When the status information is sent together with the sensor ID and the worker ID, the status information database 32 stores the status information, the sensor ID, and the worker ID in association with time information.

[0067] In this way, the work content analysis device 10 can accumulate sensing information transmitted from the sensor 110 (including the camera 115) as status information. However, even if the work content analysis device 10 is normal, there are cases where the sensing information is not transmitted due to a communication error from the sensor 110 (including the camera 115), and the status information cannot be acquired. Conversely, there are cases where, even if there is no problem with communication from the sensor 110 (including the camera 115), there are cases where the work content analysis device 10 has some kind of trouble, and therefore the sensing information cannot be received and the status information cannot be acquired.

[0068] The determination module 27 determines whether or not the work content analysis device 10 is operating normally. The determination module 27 can determine whether or not the work content analysis device 10 is operating normally by, for example, monitoring the operation of the CPU 12 and the communication unit 15.

[0069] When the state information cannot be acquired, if the determination module 27 determines that the work content analysis device 10 is normal, the processing module 21 stores first fixed data (for example, "0 (zero)") as the state information in the state information database 32, and if the determination module 27 determines that the work content analysis device 10 is not normal, the processing module 21 stores second fixed data (for example, "-1") as the state information in the state information database 32. Note that instead of "0 (zero)", a character string such as "NULL" may also be used.

[0070] By storing such fixed data in the state information database 32, the following becomes possible.

[0071] For example, a case will be described where the status information is sensing information indicating the operating status of the equipment 116. Conventionally, when sensing information indicating "1", indicating an operating status, is received from a sensor monitoring the equipment 116, "1" is stored as the status information in the status information database 32. This allows the work content analysis device 10 to determine that the equipment 116 is operating.

[0072] However, this means that status information will not be stored unless sensing information is received. Therefore, even if the work content analysis device 10 is normal, if it is unable to receive sensing information due to a communication error from the sensor monitoring the equipment 116, or conversely, if there is no problem with communication from the sensor but it is unable to receive sensing information due to a malfunction of the work content analysis device 10, status information will not be stored.

[0073] FIG. 3A is an example showing, in a time series manner, status information (sensing information indicating the operating status of a certain facility) conventionally stored in the status information database 32 in such a case.

[0074] For example, the first four times no status information was stored, so the status information is shown as (none), then for the next three consecutive times status information was received, so a "1" was stored in the status information database 32, then (none), then a "1" was stored, then (none).

[0075] In this way, when status information is not stored in the status database 32, it is completely unclear whether the cause is a problem with the work content analysis device 10 or a problem with something other than the work content analysis device 10.

[0076] In contrast, in this embodiment, when the determination module 27 determines that the work content analysis device 10 is normal but the status information cannot be acquired, the processing module 21 stores "0 (zero)" as the status information in the status information database 32. On the other hand, when the determination module 21 determines that the work content analysis device 10 is not normal but the status information cannot be acquired, the processing module 21 stores "-1" as the status information in the status information database 32. Note that instead of "0 (zero)", a character string such as "NULL" may be used.

[0077] FIG. 3B is an example showing, in time series, state information (sensing information indicating the operating state of the equipment) stored in the state information database in this embodiment.

[0078] As shown in FIG. 3B, in this embodiment, there is no (absent). Moreover, when comparing FIG. 3B with FIG. 3A, the first two (absent)s shown in FIG. 3A are shown as "-1" in FIG. 3B. This shows that the work content analysis device 10 is not normal and therefore unable to acquire status information. Moreover, the other four (absent)s shown in FIG. 3A are shown as " 0 This indicates that the work content analysis device 10 is in a normal state but is not able to acquire status information.

[0079] As another example of processing, the processing module 21 obtains second state information by performing mathematical operations on one or more pieces of first state information among the plurality of pieces of state information, and then stores the first state information and the second state information in the state information database 32. In this manner, one piece of state information can be derived from another piece of state information.

[0080] As the mathematical calculation process, for example, averaging or majority voting can be adopted. As an example of adopting averaging or majority voting as the mathematical calculation process, a case where it is applied to sensing information indicating position information will be described below.

[0081] FIG. 4A is a diagram showing an example of a time-series change in sensing information indicating an area number as position information.

[0082] In the example shown in Fig. 4A, ten pieces of sensing information indicating position information detected every second by the sensor 110 attached to the worker 100 are arranged in chronological order. The numbers indicate the area numbers. That is, Fig. 4A shows the area numbers in which the worker 100 was present for 10 seconds, every second. According to this example, the worker 100 is in area 1 for the first second, moves to area 2 after one second, returns to area 1 after another second, and thereafter moves rapidly between areas 1, 2, and 3 every second.

[0083] However, it is known that erroneous detection often occurs in detecting a position. In addition, in reality, it is unlikely that the position of the worker 100 will move between the three areas in a short time. Therefore, when handling sensing information indicating position information as status information, it is preferable to determine the status information by performing calculation processing such as averaging the sensing information or taking a majority vote within an arbitrary fixed period such as 10 seconds, rather than using raw data as is.

[0084] FIG. 4B is a diagram showing an example of state information determined by performing arithmetic processing on the sensing information shown in FIG. 4A.

[0085] In the 10-second period shown in the example of FIG. 4A, the location information indicates area 1 six times, which is the most common. Therefore, in a case such as 1 → 2 → 1, the previous and next areas 1 are judged to be correct and are confirmed. On the other hand, the intermediate area 2 is not realistic and it would be appropriate to consider it as area 1, so it is corrected to 1 → 1 → 1. Also, in a case such as 1 → 2 → 3 → 1, the first and last areas 1 are judged to be correct and are confirmed. Then, the intermediate areas 2 and 3 are also not realistic and it would be appropriate to consider them as area 1, so they are regarded as area 1 and are corrected to 1 → 1 → 1 → 1.

[0086] By such calculation processing, time-series state information as shown in FIG. 4B is obtained from sensing information indicating a time series as shown in FIG. 4A.

[0087] Although not shown, similarly to FIG. 4A, under conditions in which location information most often indicates area 1, if the location information indicates a time sequence such as 1→2→2→1, the first and last area 1 is determined and confirmed as correct, but if area 1 and area 2 are close to each other, it is possible for the person to move from area 1 to area 2, stay there for 2 seconds, and then return to area 1, so the order can simply be 1→2→2→1.

[0088] 1, the estimation module 22 estimates the work content performed by the worker 100 from at least two pieces of state information associated with the same time among the state information of the worker 100 stored in the state information database 32. Note that in this specification, the work content includes not only useful work and incidental work, such as manual work and moving work such as moving a cart, but also non-useful work such as simply walking, standing still in place, and taking a break.

[0089] For example, when estimating the work content of the worker 100(#1), the estimation module 22 targets the status information associated with the worker ID of the worker 100(#1) among the status information stored in the status information database 32. Furthermore, based on the time information associated with the status information and the sensor ID associated with the status information, the estimation module 22 focuses on at least two status information associated with the same time, and estimates the work content performed by the worker 100(#1). Note that the at least two status information associated with the same time can be not only status information of one worker, such as the worker 100(#1) in this example, but also status information of multiple workers, such as the worker 100(#1) in this example and another worker 100(#2).

[0090] As an example of a case where the at least two pieces of status information are two pieces of status information from one worker (e.g., worker 100 (#1)), a case where the pieces of status information are position information from a position sensor of one worker and acceleration information from an acceleration sensor is described. In this case, if the position information indicates within the work area and the acceleration information indicates a value greater than a predetermined value (a state in which the hands are moving for manual work), the estimation module 22 can estimate that the work content of this worker 100 (#1) is manual work.

[0091] On the other hand, if the position information does not change at all and the acceleration information indicates a state in which the acceleration is not greater than a predetermined value (a state in which the hand is barely moving), the estimation module 22 can estimate that the worker 100(#1) is stationary, that is, not doing any work. In this case, by further considering the time information, if it is within the break time of the worker 100(#1), it can be estimated that the worker 100(#1) is taking a break, and if it is not within the break time, it can be estimated that the worker 100(#1) is waiting.

[0092] As another example of a case where the at least two pieces of status information are two pieces of status information from one worker (e.g., worker 100(#1)), if the position information changes over time and the acceleration information does not change, the estimation module 22 can estimate that the worker 100(#1) was moving. However, this does not allow for distinction between whether the worker 100(#1) was moving due to work such as pushing a dolly, or whether the worker was simply walking without working. In such a case, the estimation module 22 further focuses on the sound from the microphone as status information, and can distinguish whether the worker 100(#1) was moving a dolly or simply walking depending on whether or not a sound generated by the movement of the dolly is detected.

[0093] As an example of a case where at least two pieces of status information are from multiple workers (for example, two workers, worker 100(#1) and worker 100(#2)), a case where the position information from the position sensor and the voice information from the microphone of worker 100(#1) and the position information from the position sensor and the voice information from the microphone of worker 100(#2) will be described. In this case, if the position information of worker 100(#1) and the position information of worker 100(#2) indicate that they are in the same work area, and the voice information of worker 100(#1) and the voice information of worker 100(#2) indicate that worker 100(#1) and worker 100(#2) are having a conversation about work, the estimation module 22 can estimate that worker 100(#1) and worker 100(#2) have been working together, and can further estimate the work content based on the content of the conversation.

[0094] As another example when the at least two pieces of status information are from multiple workers (for example, two workers, worker 100 (#1) and worker 100 (#2)), the pieces of status information can be at least one of position information of worker 100, movement information indicating whether worker 100's hands are moving, and information indicating the operating status of equipment around worker 100.

[0095] By using both location information and movement information, it is possible to estimate the work content on a work site basis.

[0096] By using two pieces of data, the location information and the equipment information, it is possible to estimate the behavior of the worker 100 according to the operating status.

[0097] By using two pieces of data, the movement information and the equipment information, it is possible to estimate the work of the worker 100 in accordance with the operating state.

[0098] Furthermore, for the above three cases, by using data on a group consisting of multiple workers 100, it is possible to estimate the work of the group.

[0099] The above is merely an example, and the estimation module 22 can estimate various work contents from various other combinations of state information. Then, the estimated work contents are stored in the state information database 32 in association with the corresponding state information.

[0100] The identification module 23 identifies the work time spent on the work content estimated by the estimation module 22 from the state information of the worker 100 stored in the state information database 32 and the time information associated with the state information. For example, if the estimation module 22 estimates that manual work was being performed, the identification module 23 identifies the time period from the time corresponding to the time information corresponding to the start timing of the manual work to the time corresponding to the time information corresponding to the end timing of the manual work as the work time spent on the manual work. The identification module 23 then associates the work time with the corresponding state information and stores it in the state information database 32.

[0101] The task time comparison module 24 compares the task time identified by the identification module 23 with the reference time stored in the reference time database 33 for the task content estimated by the estimation module 22. This comparison result can be the difference between the task time and the corresponding reference time. For example, if the task content is estimated as manual work by the estimation module 22, the task time identified as this manual work by the identification module 23 is 30 minutes, and 33 minutes is stored as the reference time for manual work in the reference time database 33, the task time comparison module 24 obtains a comparison result that this manual work is the reference time + 3 minutes. Alternatively, the comparison result can be a percentage of the reference time (in this case, the reference time + 10%) instead of a difference.

[0102] The analysis module 25 uses the comparison results from the work time comparison module 24 and the information stored in the state information database 32 and the reference time database 33 to analyze the work content of the worker 100 from various perspectives.

[0103] The display control module 26 causes graphs and the like showing various analysis results performed by the analysis module 25 to be displayed on the display unit 16, and also causes the external terminal 130 to display them via the communication network 70. These graphs include, for example, the results of the comparison of work times obtained by the work time comparison module 24, and work performance charts created based on the time period identified by the identification module 23.

[0104] Specific examples of the analysis performed by the analysis module 25 and the analysis results displayed by the display control module 26 will be described later.

[0105] Next, an example of the operation of the work content analysis device to which the work content analysis method of the first embodiment configured as above is applied will be described.

[0106] FIG. 5 is a flowchart showing an example of the operation of a work content analysis device to which the work content analysis method of the first embodiment is applied.

[0107] In order to analyze the work content of the workers 100, it is necessary to attach a sensor 110 to each worker 100. The number of sensors 110 attached to each worker 100 is not limited to one type, but may be multiple types.

[0108] These sensors 110 are IoT devices for visualizing the movements of the worker 100, and include, for example, a position sensor that measures the position of the worker 100 via a GPS function, a WiFi function, a Bluetooth function, or a beacon 125, an acceleration sensor that detects the hand movements of the worker 100, a vital sensor that measures vital information such as the pulse of the worker 100, a microphone that detects the voice of the worker 100, a sound sensor that detects sounds around the worker 100, a camera that captures images in the line of sight of the worker 100, and the like, but are not limited to these.

[0109] In addition, in order to analyze the work content of the worker 100, video information from a camera 115 installed at a fixed point and sensing information from a sensor attached to the equipment 116 can also be used.

[0110] Each sensor 110 is assigned a unique sensor ID and is also associated with the worker ID of the worker 100 to which it is attached. Each camera 115 is assigned a unique sensor ID.

[0111] Sensing information measured by each sensor attached to the worker 100 or equipment 116 is transmitted from the sensor 110 to the work content analysis device 10 via the communication network 70 together with the sensor ID and the worker ID. Video information, which is sensing information measured by the camera 115, is transmitted to the work content analysis device 10 via the communication network 70 together with the sensor ID (S1).

[0112] The sensing information transmitted to the work content analysis device 10 is received by the communication unit 15 together with the sensor ID and, if present, the worker ID, and is sent from the communication unit 15 to the processing module 21. The processing module 21 determines whether or not processing of the sensing information is required (S2).

[0113] If the sensing information requires processing (S2: Yes), the processing module 21 performs necessary processing on the sensing information, and acquires status information from the sensing information (S3).

[0114] In step S2, an example of sensing information that needs to be processed is video information from camera 115. In processing this video information, in processing module 21, for example, using AI technology, the worker 100 being imaged is identified from the video information, the worker ID of the worker 100 is identified, movement information and position information of the identified worker 100 are detected, and the detected movement information and position information are treated as status information.

[0115] Thus, when processing video information, the processing module 21 can function specifically as a video processing module.

[0116] In addition, the processing module 21 can also process or modify sensing information from the acceleration sensor using, for example, AI technology. Note that sensing information that needs to be processed by the processing module 21 is not limited to these.

[0117] This status information is sent from the processing module 21 to the status information database 32 together with the sensor ID and the worker ID (S4).

[0118] Examples of sensing information that requires processing include image information from the camera 115 and sensing information from an acceleration sensor.

[0119] When processing is performed on the video information from the camera 115, the processing module 21 functions as a video processing module. Then, in the processing module 21, for example, by AI technology, the worker 100 being imaged is identified from the video information, the worker ID of the worker 100 is identified, movement information and position information of the identified worker 100 are detected, and the detected movement information and position information are treated as status information.

[0120] Furthermore, when processing is performed on sensing information from the acceleration sensor, the processing module 21 acquires state information by performing processing or manipulation using, for example, AI technology.

[0121] On the other hand, if the sensing information does not require processing (S2: No), the processing module 21 does not process the sensing information, and the sensing information is sent as status information to the status information database 32 together with the sensor ID and worker ID (S5).

[0122] In the status information database 32, the status information, the sensor ID, and the worker ID sent in steps S4 and S5 are stored in association with time information (S6).

[0123] The determination module 27 constantly monitors the operation of the CPU 12 and the communication unit 15, and constantly determines whether or not the work content analysis device 10 is operating normally. In response to this, if the determination module 27 determines that the work content analysis device 10 is operating normally but the status information cannot be acquired, the processing module 21 stores "0" as the status information in the status information database 32, whereas if the determination module 27 determines that the work content analysis device 10 is not operating normally but the status information cannot be acquired, the processing module 21 stores "-1" as the status information in the status information database 32, as exemplified in Fig. 3B.

[0124] Furthermore, as described above with reference to Figures 4A and 4B, calculations are performed by the processing module 21 to determine status information, if necessary.

[0125] Thereafter, the estimation module 22 estimates the work content performed by the worker 100 from at least two pieces of state information associated with the same time among the state information of the worker 100 stored in the state information database 32 (S7). The estimation result is associated with the corresponding state information and stored in the state information database 32.

[0126] In the identification module 23, the work time spent on the work content estimated by the estimation module 22 is identified from the status information of the worker 100 stored in the status information database 32 and the time information associated with the status information (S8).

[0127] In the task time comparison module 24, for the task content estimated by the estimation module 22, the task time identified by the identification module 23 is compared with the reference time stored in the reference time database 33 (S9). The result of this comparison can be the difference between the task time and the corresponding reference time, or the ratio (percentage) of the task time to the reference time. For example, if the estimation module 22 estimates manual work as the task content, the identification module 23 identifies this manual work as having a task time of 30 minutes, and the reference time database 33 stores 33 minutes as the reference time for manual work, the task time comparison module 24 obtains a comparison result of this manual work being the reference time + 3 minutes, or the reference time + 10%.

[0128] In the analysis module 25, using the results of this comparison and the information stored in the state information database 32 and the reference time database 33, the work content of the worker 100 is analyzed from various perspectives (S10).

[0129] The results of the various analyses performed by the analysis module 25 are displayed, for example, in the form of graphs on the display unit 16 by the display control module 26 and / or on the external terminal 130 (S11). These graphs include, for example, not only the work time comparison results obtained by the work time comparison module 24 and work performance charts created based on the time period identified by the identification module 23, but also graphic displays as described below.

[0130] FIG. 6 is an example of a graphic display showing the time required for each task.

[0131] Here, as an example, it is assumed that the reference time database 33 stores the reference times for the main task, the auxiliary task, and other tasks (150 minutes for the main task, 90 minutes for the auxiliary task, and 60 minutes for the other tasks).

[0132] In addition, the status information database 32 stores the status information of the worker 100 in association with the work content (main work, auxiliary work, etc.) estimated by the estimation module 22 and the work time (120 minutes, 150 minutes, 120 minutes) for each work content identified by the identification module 23.

[0133] FIG. 6(a) shows the cumulative time (minutes) for each task, divided into reference time and actual time.

[0134] As described above, the reference time database 33 stores 150 minutes, 90 minutes, and 60 minutes as the reference times for main tasks, auxiliary tasks, and other tasks. Therefore, by referring to the reference time database 33, the analysis module 25 creates a graph in which the reference times are accumulated by task type, as shown in the reference time graph a1.

[0135] Similarly, the status information database 32 stores status information of a certain worker 100 in association with the task content (main task, auxiliary task, etc.) and the task time for each task content (120 minutes, 150 minutes, 120 minutes), so that the analysis module 25 refers to the status information database 32 to create a graph showing the accumulated task time for each task content for the worker 100, as shown in the actual time graph a2. Also, next to each task content in the actual time graph a2, the difference time (in minutes) from the same task content in the reference time graph a1 is shown.

[0136] According to this example, it is shown that the actual time of worker 100 is 30 minutes shorter than the reference time for the main task, 60 minutes longer than the reference time for the auxiliary task, and 90 minutes longer than the reference time for the other tasks.

[0137] On the other hand, FIG. 6(b) shows the time ratio for each task as a percentage.

[0138] The reference time graph b1 shows the reference time graph a1, and the performance time graph b2 shows the performance time graph a2, each showing the time ratio of each task content as a percentage, with the total time being 100%.

[0139] From the graphic displays of Figures 6(a) and 6(b), we can see that the actual time for main work is shorter than the reference time and its percentage of the total work is low, while the actual time for incidental work and other tasks is longer than the reference time, and its percentage of the total work is also higher than the reference time. Therefore, it is possible to clarify the specific countermeasures for improving work efficiency, namely, to spend more time on main work and to spend less time on incidental work and other tasks.

[0140] This kind of graphic display makes it possible to visualize the work content and identify hidden waste inherent in the work, thereby clarifying specific improvements to improve work efficiency and enabling efficient formulation of improvement plans.

[0141] Furthermore, if this improvement results in increased productivity, it can also lead to reduced production and procurement costs, resulting in cost savings.

[0142] FIG. 7 is a flowchart for explaining the flow of business improvement.

[0143] As described above, according to the work content analysis device to which the work content analysis method of the first embodiment is applied, as shown in the flowchart of FIG. 7, by acquiring sensing information from sensors 110 attached to workers 100 or equipment 116, and cameras 115 installed at fixed points (S21), various status information related to the work is collected (S22), stored in status information database 32 (S23), and various analyses are performed (S24). Based on the results of this analysis, the work content is visualized and hidden waste inherent in the work is identified, making it possible to devise specific and effective improvement plans for improving productivity (S25).

[0144] In this way, the work content analysis device 10 can analyze work content without requiring personnel to acquire sensing information and without requiring modifications to the equipment on the factory side.

[0145] If the implementation of the improvement plan (S26) results in a measured improvement in productivity (S27), it will also lead to reduced production and procurement costs, and cost reductions can also be expected.

[0146] Second Embodiment A task content analysis device to which a task content analysis method according to the second embodiment is applied will be described.

[0147] There is a risk that analyzing the work content of individual workers may lead to personal attacks. In that case, the individual may lose motivation to be productive, and it may be difficult for them to readily accept the results of the work content analysis, which may make it difficult to improve productivity.

[0148] On the other hand, if the evaluation is done as a group rather than as an individual, the results of the work analysis are more likely to be accepted because they will not be seen as personal attacks. This will also increase the sense of cooperation within the group, which is expected to lead to increased productivity.

[0149] A group may include multiple workers organized as a work team, multiple workers performing the same work, such as inspection work or assembly work, or multiple workers differentiated by role, such as team leader or inspector.

[0150] For this reason, a work content analysis device to which the work content analysis method of the second embodiment is applied performs work content analysis in groups, such as work teams.

[0151] Therefore, in this embodiment, each worker belongs to one of the groups classified in advance, and unlike the first embodiment, the work content analysis is performed on a group basis, rather than on a worker basis.

[0152] The configuration of a work content analysis device to which the work content analysis method of the second embodiment is applied can be explained using FIG. 1. Therefore, in the following, the differences from the first embodiment will be explained without repeating explanations, while referring to FIG. 1.

[0153] In this embodiment, each worker 100 is classified into one of the groups. A group can be, for example, a work group that performs work together.

[0154] Accordingly, the worker ID of each worker 100 includes not only identification information for identifying the worker 100, but also group information for specifying the group to which the worker 100 belongs. For example, if the group information is a group number given to the group in advance, the worker ID can be numeric information in which the employee number and the group number are serially combined.

[0155] Therefore, the worker database 31 stores a worker ID including identification information for identifying the worker and identification information for specifying a group to which the worker belongs, for each worker 100. Similarly, the state information stored in the state information database 32 together with the time information is also stored in association with such a worker ID.

[0156] The estimation module 22 estimates the work content performed by the group from at least two pieces of status information associated with the same time among the status information of multiple workers 100 belonging to the same group stored in the status information database 32.

[0157] For example, as an example of a case where at least two pieces of status information are from two workers 100(#1) and 100(#2) belonging to the same group, a case will be described in which the position information from the position sensor and the voice information from the microphone of the worker 100(#1) and the position information from the position sensor and the voice information from the microphone of the worker 100(#2) are included. In this case, if the position information of the worker 100(#1) and the position information of the worker 100(#2) both indicate that they are within the work area, and the voice information of the worker 100(#1) and the voice information of the worker 100(#2) indicate that the workers 100(#1) and 100(#2) are having a conversation about work, the estimation module 22 can estimate that the group was working, and can further estimate the work content based on the content of the conversation.

[0158] The above is merely an example, and the estimation module 22 can estimate various work contents performed by the group from a combination of various other status information from multiple workers belonging to the same group. Then, the estimation result is stored in the status information database 32 in association with the status information of the corresponding worker 100.

[0159] The identification module 23 identifies the task time corresponding to the task content estimated by the estimation module 22 from the status information of multiple workers 100 belonging to the same group stored in the status information database 32. For example, if the estimation module 22 estimates that manual work was being performed in the group, the identification module 23 identifies the time period from the time corresponding to the time information corresponding to the start timing of the manual work to the time corresponding to the time information corresponding to the end timing of the manual work as the task time spent on the manual work. The identification module 23 then associates the task time with the status information of the corresponding worker 100 and stores it in the status information database 32.

[0160] As in the first embodiment, the work time comparison module 24 compares the work time identified by the identification module 23 for the work content estimated by the estimation module 22 with the reference time stored in the reference time database 33.

[0161] The analysis module 25 analyzes the work content of the group based on the work content estimated by the estimation module 22 and the work time identified by the identification module 23. Alternatively, the analysis module 25 analyzes the work content of each worker 100 belonging to the group based on the work content estimated by the estimation module 22 and the work time identified by the identification module 23, and further analyzes the work content of the group as a whole based on the results of this analysis.

[0162] Next, an example of the operation of the work content analysis device to which the work content analysis method of the second embodiment configured as above is applied will be described.

[0163] FIG. 8 is a flowchart showing an example of the operation of a task content analysis device to which the task content analysis method of the second embodiment is applied.

[0164] In FIG. 8, the same processes as those in FIG. 5 are denoted by the same step numbers, and only the different contents will be explained to avoid duplication of explanation.

[0165] As in the first embodiment, one or more types of sensors 110 are attached to each worker 100. Also, as in the first embodiment, a camera 115 is installed at a fixed point.

[0166] Each sensor 110 is assigned a sensor ID and associated with a worker ID, and the worker ID includes not only the identification information of the worker 100 but also the identification number of the group to which the worker 100 belongs.

[0167] Therefore, in the following description of FIG. 5, it is assumed that the worker ID includes not only the identification information of the worker 100 but also the identification number of the group to which the worker 100 belongs.

[0168] In the estimation module 22, the work content performed by the group is estimated from at least two pieces of status information associated with the same time among the status information of multiple workers 100 belonging to the same group stored in the status information database 32 (S7a).

[0169] The identification module 23 identifies the task time spent on the task content estimated by the estimation module 22 from the status information of multiple workers 100 belonging to the same group stored in the status information database 32 (S8a).

[0170] In the work time comparison module 24, as in the first embodiment, the work time identified by the identification module 23 for the work content estimated by the estimation module 22 is compared with the reference time stored in the reference time database 33 (S9a).

[0171] In the analysis module 25, the work content of the group is analyzed based on the work content estimated by the estimation module 22 and the work time identified by the identification module 23. Alternatively, the work content of each worker 100 belonging to the group is analyzed based on the work content estimated by the estimation module 22 and the work time identified by the identification module 23, and further, the work content of the group as a whole is analyzed based on this result (S10a).

[0172] This analysis can be performed from various viewpoints using the information stored in the state information database 32 and the reference time database 33, similarly to the first embodiment.

[0173] A graph showing the results of various analyses performed by the analysis module 25 on the group is displayed on the display unit 16 or the external terminal 130 by the display control module 26 (S11a).

[0174] Below, examples of graphic displays that are displayed on the display unit 16 and / or the external terminal 130 based on the analysis results by the analysis module 25 will be presented.

[0175] FIG. 9 is an example of a graphic display showing the frequency and duration of visits to each area of ​​a factory by all workers in a group.

[0176] Fig. 9(a) shows the cumulative time that worker 100 spent in each area A to J performing manual work in one day (or one week). In Fig. 9(a), the vertical axis represents the cumulative time, and the horizontal axis represents the area.

[0177] According to Fig. 9(a), it can be seen that the workers belonging to the group stayed in Area D for a cumulative 6 hours and 15 minutes for manual work. Furthermore, according to Fig. 9(b), it can be seen that the breakdown of the 6 hours and 15 minutes is that Worker X stayed for 3 hours and 33 minutes, Worker Y stayed for 1 hour and 24 minutes, and Worker Z stayed for 1 hour and 18 minutes.

[0178] FIG. 9(c) shows the result of three classifications based on the ratio of stay time in each area to the total cumulative time in all areas shown in FIG. 9(a) (area concentration). In FIG. 9(c), each area is shown in a schematic diagram according to the positional relationship according to the actual layout in the factory. According to the example in FIG. 9(c), only area D has an area concentration of 21% or more due to manual work, only area I has an area concentration of 4% or less, and the other areas have area concentration of 5-20%.

[0179] As shown in the legend in FIG. 9(a), it is also possible to display the display as in FIG. 9 according to the type of work performed by the worker 100, such as not only manual work but also cart work, walking, and stationary work.

[0180] This type of display makes it possible to visualize the actual on-site behavior of the entire group, such as which areas workers tend to concentrate in when doing which tasks, and which personnel have a heavy or light burden.

[0181] FIG. 10 shows another example of a comparison of the work performance of a person with a high level of skill and a person with a low level of skill.

[0182] Figure 10 is an example of a graphic display showing the actual work hours broken down by work type.

[0183] In this example, the work contents are classified into main work, auxiliary work, and others.

[0184] In FIG. 10(a), the vertical axis represents cumulative time, and in FIG. 10(b), the vertical axis represents time percentage (%).

[0185] 10(a) and 10(b), which are comparisons of the actual working hours between a highly skilled person, such as an experienced worker, and a less skilled person, such as a beginner, it can be seen that compared to the experienced worker X, the beginner worker Y has a longer total working time, a lower proportion of main work, and a higher proportion of incidental work and other work. Therefore, it can be specifically clarified that the improvement item to improve the work efficiency of beginners is to reduce the proportion of incidental work and other work and increase the proportion of main work, thereby shortening the total working time.

[0186] As described above, according to the work content analysis device to which the work content analysis method of the second embodiment is applied, the work content analysis performed for each individual worker in the first embodiment can be performed on a group basis.

[0187] This means that workers will no longer be criticized individually, and each worker will be more likely to accept the analysis results and the items for improvement that are derived from them. This will inevitably encourage each worker to proactively work on improving their work, and we can expect to foster a sense of cooperation within the group. This can be expected to revitalize the group, for example by increasing communication within the group and encouraging workers to proactively work on necessary work improvements, such as changing work procedures and layouts.

[0188] And ultimately, it is highly expected that this will lead to cost reduction effects such as improved productivity and reduced production and procurement costs.

[0189] In the above, we have explained how to improve work efficiency by focusing on the work time and the breakdown of work time of experienced workers. However, it is also possible to focus on the locations of experienced workers (for example, the experienced worker went to a certain location but the novice worker did not) or the routes they took (the experienced worker took this route but the novice worker took a different route) and use these to improve work efficiency.

[0190] Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope of the invention and its equivalents described in the claims, as well as in the scope and spirit of the invention. [Explanation of symbols]

[0191] 10... Work content analysis device, 11... Bus, 13... External recording medium, 14... Recording medium reading unit, 15... Communication unit, 16... Display unit, 20... Memory, 21... Processing module, 22... Estimation module, 23... Identification module, 24... Work time comparison module, 25... Analysis module, 26... Display control module, 27... Judgment module, 29... Writable data area, 30... Storage device, 31... Worker database, 32... Behavioral information database data, 33··reference time database, 70··communication network, 100··worker, 110··sensor, 115··camera, 116··equipment, 120··microcontroller, 125··beacon, 130··external terminal, 1101··bus, 1102··CPU, 1103··sensing unit, 1104··transmitting unit, 1105··receiving unit, 1110··memory, 1111··communication determination module, 1120··storage device, 1121··sensing information database

Claims

1. A work content analysis device that analyzes work content of a worker based on a state of the worker, a first database that stores status information representing the status of one or more of the workers in association with time information and identification information of the workers, the status information including position information of the workers, movement information indicating whether the workers' hands are moving, and information indicating the operating status of equipment around the workers; an estimation unit that estimates the work content performed by the worker based on state information associated with the work content and stored in the first database at the same time; an identification unit that identifies a task time spent on the estimated task content from the state information stored in the first database and time information associated with the state information; an analysis unit that analyzes the work content based on the estimated work content and the specified work time; a determination unit that determines whether the task content analysis device is operating normally; a processing unit that stores, in the first database, first fixed data as the status information if the determining unit determines that the work content analysis device is normal when the status information cannot be acquired, and stores, in the first database, second fixed data as the status information if the determining unit determines that the work content analysis device is not normal. A work content analysis device comprising:

2. 2. The work content analysis device according to claim 1, further comprising a processing unit that obtains second status information by performing mathematical arithmetic processing on one or more pieces of first status information among the plurality of pieces of status information, and stores the first status information and the second status information in the first database.

3. The task content analysis device according to claim 1 , wherein the state information is acquired by a sensor attached to the worker or a sensor installed around the worker, and is stored in the first database.

4. 4. The work content analysis device according to claim 3, wherein the first database stores the sensor and identification information of a worker to whom the sensor is attached, or identification information of a worker within a predetermined distance from a location where the sensor is installed, in association with each other.

5. A second database that stores a reference time determined in advance for each of the work contents; The task content analyzing device according to claim 1 , further comprising a task time comparing unit that compares the identified task time with the corresponding reference time for the estimated task content.

6. The task content analysis device according to claim 5 , wherein the comparison result includes a difference or a ratio between the task time and the corresponding reference time.

7. The task content analysis device according to claim 6 , further comprising a display control unit that displays the difference or the ratio for each task content.

8. The task content analysis device according to claim 6 , wherein the analysis unit further analyzes the task content of the worker based on the difference or the ratio.

9. The task content analysis device according to claim 1 , further comprising a display control unit that displays a result of the analysis by the analysis unit.

10. The identification unit further identifies a time period corresponding to the estimated work content from the state information stored in the first database, The work content analysis device according to claim 9 , wherein the display control unit displays, as the analysis result, a work performance diagram in which the work content is displayed on a time axis based on the specified time period.

11. A work content analysis device that analyzes work content of a worker based on the worker's state, comprising: a first database that stores status information representing the status of one or more of the workers in association with time information and identification information of the workers, the status information including position information of the workers, movement information indicating whether the workers' hands are moving, and information indicating the operating status of equipment around the workers; an estimation unit that estimates the work content performed by the worker based on state information associated with the work content and stored in the first database at the same time; an identification unit that identifies a task time spent on the estimated task content from the state information stored in the first database and time information associated with the state information; an analysis unit that analyzes the work content based on the estimated work content and the specified work time, The worker belongs to any one of the groups previously classified, the identification information includes group information specifying a group to which the worker belongs, the estimation unit estimates the work content performed by the group based on status information stored in the first database that is associated with the same time among status information of workers belonging to the same group, and The identification unit identifies a task time spent on the estimated task content from status information of workers belonging to the same group stored in the first database, and The analysis unit analyzes the work content of the group based on the estimated work content and the specified work time.

12. 2. The work content analysis device according to claim 1, wherein the status information further includes at least any of movement information indicating whether the worker is moving or not, vital sign information of the worker, voice information of the worker, ambient sound information of the worker, and video information of the worker's surroundings.

13. 2. The work content analysis device according to claim 1, wherein the time information stored in the first database is synchronized with a clock means of a system to which the work content analysis device itself is connected.

14. The task content analysis device according to claim 1 , wherein the task content estimated by the estimation unit includes a break for a worker.

15. A work content analysis method implemented by a work content analysis device for analyzing work content of a worker based on a state of the worker, comprising: The work content analysis device includes: storing status information representing the status of one or more of the workers in a database in association with time information and identification information of the workers, the status information including position information of the workers, movement information indicating whether the hands of the workers are moving, and information indicating the operating status of equipment around the workers; estimating the work content performed by the worker based on state information associated with the same time and stored in the database; Identifying a task time spent on the estimated task content from the state information stored in the database and time information associated with the state information; Analyzing the work content based on the estimated work content and the specified work time; Determining whether the task content analysis device is operating normally; When the status information cannot be obtained, if it is determined that the work content analysis device is normal, first fixed data is stored in the database as the status information, and if it is determined that the work content analysis device is not normal, second fixed data is stored in the database as the status information.

16. A program applied to a work content analysis device that analyzes work content of a worker based on the worker's state, a function of storing status information representing the status of one or more of the workers in a database in association with time information and identification information of the workers, the status information including position information of the workers, movement information indicating whether the workers' hands are moving, and information indicating the operating status of equipment around the workers; A function of estimating the work content performed by the worker based on state information associated with the same time and stored in the database; a function of identifying a task time spent on the estimated task content from the state information stored in the database and time information associated with the state information; A function of analyzing the work content based on the estimated work content and the specified work time; A function of determining whether the work content analysis device is operating normally; a function of storing, in the database, first fixed data as the status information if it is determined that the work content analysis device is normal when the status information cannot be acquired, and storing, in the database, second fixed data as the status information if it is determined that the work content analysis device is not normal. A program to make the above happen on a computer.

17. A sensor for acquiring the state information to be transmitted to the task content analysis device according to claim 3, A sensing unit that acquires status information of the worker; a state information database that stores the acquired state information; a communication determination unit that determines whether communication with the task content analysis device is possible; a transmission unit that transmits, when the communication determination unit determines that communication is possible, the status information stored in the status information database to the work content analysis device in predetermined units.

18. The sensor according to claim 17 , further comprising a radio wave receiving unit that receives radio waves from a radio wave transmitter whose radio wave intensity is variable.

19. 19. The sensor of claim 17 or 18, attached to the worker or located in the vicinity of the worker.

20. The sensor of claim 19, wherein when attached to the worker, the sensor is attached to any one of the front of the worker's body, the left or right side of the waist, the arm, the wrist, the neck, the shoulder, the ear, and the head.

21. The sensor of claim 20 , wherein the sensor is attached to the worker's dominant arm or hand when acquiring the movement information.

Images