Production management system, production management method, and program thereof

Abstract

To provide a production management system that facilitates production management based on the turnout expected from work.SOLUTION: A production management system includes an extraction unit, an acquisition unit, and a determination unit. The extraction unit extracts a predetermined region where a commodity, which is a work product produced by a worker, is to be placed, from a video of a work site. The acquisition unit detects whether a video of the shape of a person's hand is included in a video of the predetermined region, and records a time when the shape of the person's hand is detected and a time when the detection of the shape of the person's hand becomes unavailable, as an analysis log. The determination unit adds the actual turnout of the commodity when a length of time from the time when the shape of the person's hand is detected to the time when the detection of the shape of the person's hand becomes unavailable is larger than a predetermined value.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to a production management system, a production management method, and a program thereof.

Background Art

[0002] In recent years, in domestic manufacturing sites, there is a shortage of workers and a need for small-lot, multi-variety production, and it is required to operate conventional operations more efficiently with fewer personnel. Conventionally, the grasping and improvement of operations performed manually, which are generally carried out, are methods such as production managers going to the production site and measuring the operation with a stopwatch. Since data collection, analysis, and improvement consideration are carried out manually, the operation cost is high, and it is only locally applied in scenes determined to be particularly necessary, and it has not been possible to carry out constantly.

[0003] On the other hand, for example, Patent Document 1 discloses a technique for assisting in improving a work process by detecting a hand from an image obtained by imaging a work area and determining the stop of work in the work process based on the result of the hand detection.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] The ultimate goal of improving the work process is production management to obtain the production output expected by the work. The technique for assisting in improving the work process described in Patent Document 1 is an indirect support for obtaining the production output expected by the work, but it cannot directly manage the production output.

[0006] An object of the present disclosure is to provide a production management system, a production management method, and a program thereof that solve the above problems.

Means for Solving the Problem

[0007] A production management system according to an aspect of the present disclosure includes: an extraction means for extracting a predetermined area for placing a product that is a result of work by a worker from an image of a work site; a determination means for detecting whether there is an image of a human hand shape in the image of the predetermined area, and recording, as an analysis log, the time when the human hand shape was detected and the time when the human hand shape could no longer be detected after the detection; and a management means for adding the actual production volume of the product when the elapsed time from the time when the human hand shape was detected to the time when the human hand shape could no longer be detected after the detection of the human hand is greater than a predetermined value.

[0008] A production management method according to an aspect of the present disclosure includes: extracting a predetermined area for placing a product that is a result of work by a worker from an image of a work site; detecting whether there is an image of a human hand shape in the image of the predetermined area; recording, as an analysis log, the time when the human hand shape was detected and the time when the human hand shape could no longer be detected after the detection; and adding the actual production volume of the product when the elapsed time from the time when the human hand shape was detected to the time when the human hand shape could no longer be detected after the detection of the human hand is greater than a predetermined value, which is performed by a computer executing the above.

[0009] A program for production management according to an aspect of the present disclosure causes a computer to: extract a predetermined area for placing a product that is a result of work by a worker from an image of a work site; detect whether there is an image of a human hand shape in the image of the predetermined area; record, as an analysis log, the time when the human hand shape was detected and the time when the human hand shape could no longer be detected after the detection; and add the actual production volume of the product when the elapsed time from the time when the human hand shape was detected to the time when the human hand shape could no longer be detected after the detection of the human hand is greater than a predetermined value.

Advantages of the Invention

[0010] According to the above aspect, production management based on the expected production volume by work becomes easy.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Mode for Carrying Out the Invention

[0012] Hereinafter, each embodiment will be described with reference to the drawings. In all the drawings, the same or corresponding components are denoted by the same reference numerals, and common descriptions are omitted. FIG. 1 is a diagram showing the functional blocks of the production management system in the present disclosure. The production management system (1) includes a control unit (2) and a recording unit (3). The control unit (2) includes an extraction unit (21), an acquisition unit (22), and a determination unit (23) as functions for performing production management. Further, the recording unit (3) records an analysis log (31) and management parameters (32) as information necessary for performing production management and information obtained by processing.

[0013] The extraction unit (21) of the control unit (2) extracts the placement area (5) for placing the product, which is the result of the operator's work, from the video of the work site by the operator's manual operation. The acquisition unit (22) detects whether there is an image of a human hand shape in the video of the placement area (5), and records the time when the human hand shape is detected and the time when the human hand shape can no longer be detected after the detection as an analysis log. When the elapsed time from the time when the human hand shape is detected to the time when the human hand shape can no longer be detected after the detection of the human hand is greater than a predetermined value, the determination unit (23) adds the actual production quantity of the product. In addition, the determination unit (23) compares the takt time, which is the expected production time per product, with the cycle time, which is the production time of one product actually produced by the operator, and also performs work process analysis such as whether there is a delay in the work by the operator.

[0014] The analysis log (31) of the recording unit (3) holds the data obtained as a result of the processing in the control unit (2). The management parameter (32) records the data for the processing in the extraction unit (21), the parameters for data collection by the acquisition unit (22), and the parameters for work analysis by the determination unit (23).

[0015] In addition, the production management system (1) includes a photographing device (4) that photographs the work site by the operator's manual operation. The production management system (1) and the photographing device (4) are connected via a network (6) such as a LAN (Local Area Network).

[0016] The photographing device (4) photographs the work situation of the operator's manual operation, which is the object of management and work analysis, and the product obtained by the work. In the present disclosure, as an example, there are part A and packing box B on the operator's workbench, and the operator manually packs part A in packing box B to make product C. In addition, a work site where the operator aims to produce 30 products in 1 hour of work time will be assumed for explanation. In FIG. 1, reference numeral 51 is a workbench for producing product C, and when product C, which is the result of the operator's production activity, is completed on the workbench, it is placed in a predetermined area of the desk. Here, the predetermined area for placing the result of the workbench is defined as the placement area (5).

[0017] FIG. 2 is a diagram illustrating the imaging direction of the camera (4) relative to the workbench and the placement area (5) on it. In FIG. 2, assume that a worker is performing packaging work at one end of the workbench (51) and placing the finished product in the placement area (5) at the other end. In this case, imaging direction A is easier to visually distinguish the work of placing the finished product in the placement area (5) than imaging direction B. This is because imaging direction A makes it easier to determine from the image whether the worker is placing their hand in the product area while working, compared to imaging direction B, and also makes it easier to capture images that include the workbench (51) and the worker. Therefore, the angle of view of the camera (4) is set, for example, to a range that can capture both the workbench (51), including the placement area (5), and the worker. This allows for normal worker work monitoring using the camera (4).

[0018] Next, the operation of the production management system (1) will be explained. Here, the camera (4) captures images of the work site with an angle of view that covers the workbench (51) and workers, and transmits the images in real time to the production management system (1) via the network (6). The production management system (1) receives the images, analyzes them, and manages production based on the analysis results. The mechanisms of image analysis and production management by the production management system (1) will be explained below.

[0019] FIG. 3 is a diagram showing the processing flow of video analysis by the extraction unit (21) and acquisition unit (22) of the control unit (2). As a setting before starting video analysis, a placement area (5) in the video is set (S11). For example, when a fixedly installed imaging device (4) inputs video with a resolution of 1920 x 1080 pixels (FHD), the placement area (5) can be set at any coordinate "(x1, y1)-(x2, y2)" within the coordinate range (x, y) = (0, 0) to (1920, 1080), enabling hand shape detection within that area. This set area is recorded as part of the management parameters (32) of the recording unit (3).

[0020] The production management system (1) receives (S12) the video captured from the imaging device (4) at an angle of view that includes the workbench (51) and the operator. Here, the video is composed of a plurality of consecutive images. In the video processing by the extraction unit (21), actually, it is the processing of the images that make up the video, but here it is called video processing.

[0021] The extraction unit (21) extracts only the set area included in the management parameter (32) from the received video. Then, the acquisition unit (22) performs a detection process (S13) to check whether there is a human hand shape in the extracted video. In this way, the imaging device (4) can be used in combination with the conventional surveillance camera application of photographing and storing the entire workbench (51) and the operator. Also, the acquisition unit (22) outputs (S13, S14) the time when the human hand shape is detected as the analysis log (31). Here, the time output by the acquisition unit (22) is from the time Tn(in) when the hand shape is detected to the time Tn(out) when the hand shape detection stops after the hand shape is detected. That is, the time when the operator starts to place the completed product in the placement area (5) and the time when the placement is completed are recorded. Here, the subscript n of Tn(in) and Tn(out) is a positive integer indicating the number of the time when the hand shape is detected and the time when the hand shape cannot be detected from the start of the process in the production management system (1). Note that the detection of the human hand shape is performed by using image matching or the learning result of the human hand shape by AI. The extraction unit (21) determines (S15) whether the detection of the hand shape should be continued. If continuation is necessary (S15: Yes), the process proceeds to S12. Note that the extraction unit (21) determines whether continuation is necessary, for example, based on whether the number of products created has reached the expected production quantity, which will be described later.

[0022] Next, the mechanism of production management will be described. FIG. 4 is a diagram showing the processing flow by the determination unit (23) of the control unit (2). The determination unit (23) checks (S21) the Tn(in) and Tn(out) included in the analysis log (31) of the recording unit (3) at any time.

[0023] When the determination unit 23 confirms the new Tn(in) and Tn(out), it calculates the elapsed time Tn = Tn(out) - Tn(in) from the time Tn(in) when the hand was detected to the time Tn(out) when the hand was no longer detected. Furthermore, the determination unit 23 compares the elapsed time Tn with a predetermined value Tdelta [sec], which is a production control parameter that can be set by the user of the production control system 1 (S22).

[0024] If the elapsed time Tn is greater than Tdelta, the determination unit (23) determines that the worker has completed production of the product C and placed the finished product C in the placement area (5), and measures the worker's production volume by incrementing N by +1, outputting the volume to the analysis log (31) (S23). The volume N is initialized (N = 0) when the production management system (1) starts processing. The purpose of setting the predetermined value Tdelta is to eliminate false detections caused by, for example, the worker's hand momentarily entering the placement area (5). Through the processes of S22 and S23 in FIG. 4, the acquisition unit (22) automatically detects the worker's hand through video analysis each time the worker produces the product C and places the finished product C in the placement area (5). Furthermore, the determination unit (23) processes the time data recorded every moment in the analysis log (31) and adds the worker's production volume to the analysis log (31). If the determination unit 23 determines that the elapsed time Tn is not greater than Tdelta, i.e., that the detection is erroneous, it is preferable that the determination unit 23 deletes Tn(in) and Tn(out) or makes it possible to identify them as erroneous detection data. More preferably, if the elapsed time Tn is greater than Tdelta, the determination unit 23 uses the output value N to separately store the value of the subscript n of Tn(in) and Tn(out) as the value of the output N. This is because doing so makes it possible to easily calculate the cycle time (CT), which is the time it takes for a worker to create one product, as described below. In the following processing of S25 and S26, the determination unit 23 will be described as performing processing using the separately stored Tn(in) and Tn(out) when the elapsed time Tn is greater than Tdelta.

[0025] The determination unit (23) determines whether the output volume N is greater than 1 (N>1) (S24). If the output volume N is not greater than 1 (S24: No), the determination unit (23) returns the process to S21. On the other hand, if the output volume N is greater than 1 (S24: Yes), the determination unit (23) determines the time when the worker completes production of the (N-1)th product as the start of production of the (N)th product, and calculates the cycle time (CT) from the start of production of the (N)th product to the completion of production of the (N)th product. In addition, the determination unit (23) saves the calculated cycle time (CT) in the analysis log (31) (S25).

[0026] It is assumed that the production time H and expected output M are set as user-configurable management parameters (32) in the production management system (1). The production management system (1) provides a mechanism for more advanced production management by using the production time H and expected output M in conjunction with the analysis log (31). In this example, the production time H = 1 [h] and the expected output M = 30 [units]. In this case, the worker is expected to produce 30 units of product C in one hour, so the worker is expected to work for 2 minutes per unit (= 60 minutes / 30 units). This time will be referred to as takt time (TT). In the following example, it is assumed that the worker's work will end when the expected output M is reached.

[0027] In the above case, barring erroneous determination, it is expected that the time at which a hand is detected will be repeated from T1, T2, ..., Tn (n: 1 to 30). Therefore, the determination unit (23) calculates the cycle time CTn = Tn(in) - Tn-1(out) (where the subscript n>1) for the nth product using Tn(in), which is the time at which a hand is detected in the placement area (5) when the nth product is completed, and Tn-1(out), which is the time at which a hand is no longer detected after being detected in the placement area (5) when the (n-1)th product is completed, to obtain the cycle time (CT), which is the time at which each of the second and subsequent products C is produced, and stores this in the analysis log (31) (S25).

[0028] The determination unit (23) stores the comparison result of the tact time (TT) - cycle time CTn (subscript n > 1) together with "n" which becomes the actual production quantity number of the product in the analysis log (31). Also, here, the determination unit (23) determines whether TT - CTn < 0 (S26). When TT - CTn < 0 (S26: Yes), the determination unit (23) determines that a delay has occurred in the actual manufacturing time with respect to the planned time (S27), and records to that effect. In this way, by recording the information that a delay has occurred in the actual manufacturing time with respect to the planned time, it can be determined that there is some problem in the production process by the operator of the nth product c. Furthermore, in the monitoring video, it becomes possible to easily specify the work video from the time Tn - 1(out) to the time Tn(in) which is the production time of the nth product c, and it becomes possible to easily perform work analysis of the operator based on the video obtained by the imaging device (4) imaging the workbench (51) and the operator.

[0029] On the other hand, when TT - CTn is not < 0 (S26: No), the determination unit (23) determines whether the production quantity N by the work reaches the expected production quantity M (S28). When the production quantity N does not become the expected production quantity M (S28: No), since the work of the operator continues, the determination unit (23) transfers the process to S21. On the other hand, when the production quantity N becomes the expected production quantity M (S28: Yes), since the work by the operator is completed, the production management system (1) ends the production management process.

[0030] Note that in S26, the determination unit (23) compares the tact time by using the time required to produce each one product as the cycle time. It is not limited to this, and the determination unit (23) may use the average value of the cycle times of a predetermined number of products as the cycle time and compare it with the tact time. Although the production time of each product may vary, by taking the average, it becomes possible to evaluate the work with reduced variation. Also, by making the predetermined number when taking the average match the expected production quantity M during the production time H, it also becomes possible to check the presence or absence of a delay with respect to the entire production time H. Also, in the process shown in S26, the cycle time CTn may use a moving average. Also in this case, it becomes possible to evaluate the work with reduced variation.

[0031] In addition, the judgment unit (23) may determine whether or not there is a work delay by comparing the expected number of products produced per unit time with the actual number of products produced by the operator per the same unit time, as an indicator for determining whether or not there is a work delay.

[0032] Exceeding planned production time is handled by working overtime, etc., which increases production costs at the manufacturing site, resulting in a net loss. By using the production management system (1) disclosed herein, it is possible to provide a production management mechanism that uses the imaging device (4) as a sensor and automatically measures and visualizes output and production progress delays in a simple manner. This enables early correction and countermeasures to be taken at a wide range of manual work sites, and can be used to improve productivity at production sites.

[0033] As an application example of the present disclosure, an example of production management in which each worker places their product on a conveyor belt after completing their work will be described below. FIG. 5 is a diagram showing an example in which work tables are arranged so as to be connected to multiple conveyor belts, and finished products are placed on the conveyor belts for delivery. When a worker places a product on the conveyor belt after completing work in the work area, by detecting the timing when the worker's hand is on the conveyor belt, it is possible to perform output management and progress management for multiple workers in a similar manner. In this case, the above-mentioned processing is performed for each worker, i.e., for each installed imaging device (4).

[0034] As described above, the production management system disclosed herein can utilize existing imaging devices (4) for monitoring work status, eliminating the need to select and install IoT sensors according to the work environment. Furthermore, by analyzing video footage of workers performing specific operations and generating human work data in real time, simple understanding of output and production progress becomes possible. Additionally, by sequentially comparing the cycle time with the takt time of the day, work progress can be grasped in real time. It also makes it easier to obtain information for guidance and correction for delayed workers.

[0035] In addition, by utilizing the production management system of the present disclosure, the cycle time for each worker can be obtained as quantitative results, enabling the assessment of the workers' capabilities. Since the working hours of workers with short and long cycle times are retained as data, instead of casually reviewing the recorded videos, it can be utilized for worker training by comparing the videos of specific parts.

[0036] If it changes with the normal tact time > cycle time, the priority of work improvement decreases and improvement work is often not carried out. However, due to the low introduction cost, potential improvement areas can be grasped and addressed by checking the actual values of the cycle time in a wide range of processes.

[0037] By using video analysis, it is possible to generally grasp the output and progress of manual work at the production site. Production managers can efficiently and constantly grasp the progress of multiple workers by checking the automatically output progress information. As a result, it becomes possible to take corrective measures necessary for workers with poor progress. As a result, it is possible to reduce the pure loss that requires overtime work because the work is not completed within the specified time, that is, extra time and costs are incurred in production, and improve productivity.

[0038] By providing a system that uses a camera to film the work of workers and uses video analysis to turn the camera into a sensor for manual work measurement, and automatically visualizes the output management and progress status, it can be widely used in various production sites / methods, and it is possible to effectively utilize the existing camera equipment and lead to an improvement in productivity.

[0039] FIG. 6 is a diagram showing a configuration example of a production management system (1) according to an embodiment of the present disclosure. The production management system (1) includes an extraction unit (21) (extraction means), an acquisition unit (22) (acquisition means), and a determination unit (23) (determination means). The extraction unit (21) extracts a predetermined area for placing a product that is the result of work by an operator from the video of the work site. The acquisition unit (22) detects whether there is an image of a human hand shape in the video of the predetermined area, and records the time when the human hand shape is detected and the time when the human hand shape can no longer be detected after the detection as an analysis log. When the elapsed time from the time when the human hand shape is detected to the time when the human hand shape can no longer be detected after the detection of the human hand is greater than a predetermined value, the determination unit (23) adds the actual production volume of the product.

[0040] FIG. 7 is a block diagram showing an example of the hardware configuration of the production management system (1). Here, the production management system (1) is composed of a CPU (61), a RAM (Random Access Memory) (62), a non-volatile memory (63) such as a ROM (Read Only Memory), a recording device (64), and the like. A program for realizing the functions of the control unit (2) of the production management system (1) is recorded in the non-volatile memory (63) and the recording device (64). The RAM (62) is used as a working area or the like for temporarily storing data and the like used when the CPU (61) and the like are operating. The production management system (1) may also include an input / output port (65) for connecting input / output devices such as a keyboard, a mouse, and a display device. These devices are connected by a bus or the like. Note that the non-volatile memory (63) is composed of an EEPROM (Electrically Erasable Programmable Read-Only Memory) or the like, and the recording device (64) is composed of a hard disk, an SSD, or the like, and a computer program for realizing the functions of the production management system (1) may be updated by these devices.

[0041] Although the present disclosure has been described with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. And each embodiment can be combined with other embodiments as appropriate.

[0042] Some or all of the above-described embodiments may be described as follows in the appended claims, but are not limited thereto.

[0043] (Appendix 1) Extraction means for extracting a predetermined area for placing a product that is the result of work by a worker from the video of the work site, Determination means for detecting whether there is an image of a human hand shape in the video of the predetermined area and recording, as an analysis log, the time when the human hand shape was detected and the time when the human hand shape could no longer be detected after the detection, Management means for adding the actual production output of the product when the elapsed time from the time when the human hand shape was detected to the time when the human hand shape could no longer be detected after the detection of the human hand is greater than a predetermined value, A production management system comprising:

[0044] (Appendix 2) The management means records the elapsed time from the time when the human hand shape can no longer be detected to the time when the human hand shape is detected next as a cycle time, which is the time for creating the product. The production management system according to Appendix 1.

[0045] (Appendix 3) The management means compares the cycle time, which is the expected creation time per unit of the product, with the cycle time, and determines that there is a work delay when the cycle time is longer than the tact time. The production management system according to Appendix 1 or 2.

[0046] (Appendix 4) The management means sets the average value or the moving average value of the cycle times of a predetermined number of the products as the cycle time. The production management system described in Supplementary Note 3.

[0047] (Supplementary Note 5) The management means determines the presence or absence of work delays by comparing the expected production quantity per unit time of the product with the actual production quantity per unit time. The production management system according to any one of Supplementary Notes 1 to 4.

[0048] (Supplementary Note 6) The extraction means extracts the area specified by the coordinate position in the video as the predetermined area. The production management system according to any one of Supplementary Notes 1 to 5.

[0049] (Supplementary Note 7) It is provided with a photographing device for photographing the work site including the predetermined area. The extraction means extracts the predetermined area from the video from the photographing device. The production management system according to any one of Claims 1 to 6.

[0050] (Supplementary Note 8) The photographing device is provided for each work site in order to photograph a plurality of work sites including the predetermined area. The extraction means, the determination means, and the management means each perform processing on the video obtained for each photographing device. The production management system according to Supplementary Note 7.

[0051] (Supplementary Note 11) Extract a predetermined area for placing the product that becomes the work result by the worker from the video of the work site. Detect whether there is an image of a human hand shape in the video of the predetermined area, and record the time when the human hand shape is detected and the time when the human hand shape cannot be detected after detection as an analysis log. When the elapsed time from the time when the human hand shape is detected to the time when the human hand shape cannot be detected after the detection of the human hand is greater than a predetermined value, add the actual production quantity of the product. A production management method performed by a computer executing the above.

[0052] (Appendix 12) The time elapsed from the time when the shape of the person's hand can no longer be detected to the time when the shape of the person's hand is next detected is recorded as a cycle time, which is the time taken to produce the product. 11. A production control method as described in Appendix 11.

[0053] (Appendix 13) A takt time, which is an expected production time per unit of the product, is compared with the cycle time, and if the cycle time is longer than the takt time, it is determined that there is a delay in the work. 13. The production management method according to claim 11 or 12.

[0054] (Appendix 14) The cycle time is an average or moving average value of the cycle time for a predetermined number of the products. 13. A production control method as described in Appendix 13.

[0055] (Appendix 15) determining whether or not there is a delay in the work by comparing the expected output of the product per unit time with the actual output per unit time; 15. A production management method according to any one of appendices 11 to 14.

[0056] (Appendix 16) extracting an area specified by a coordinate position in the video as the predetermined area; 16. A production management method according to any one of appendices 11 to 15.

[0057] (Appendix 17) Extracting the predetermined area from the image taken by the imaging device that takes an image of the work site including the predetermined area. 17. A production management method according to any one of appendices 11 to 16.

[0058] (Appendix 18) Performing processing on the video from the imaging device provided for each of the work sites to capture a plurality of work sites including the predetermined area, respectively. The production management method according to Supplementary Note 17.

[0059] (Supplementary Note 21) Extracting a predetermined area for placing the product that is the result of the work by the worker from the video of the work site, Detecting whether there is an image of a human hand shape in the video of the predetermined area, and recording the time when the human hand shape was detected and the time when the human hand shape could no longer be detected after the detection as an analysis log. When the elapsed time from the time when the human hand shape was detected to the time when the human hand shape could no longer be detected after the detection of the human hand is greater than a predetermined value, adding the actual output of the product. A program for production management that causes a computer to execute the above.

[0060] (Supplementary Note 22) Recording the elapsed time from the time when the human hand shape could no longer be detected to the time when the human hand shape was detected next as the cycle time, which is the time to create the product. The program according to Supplementary Note 21.

[0061] (Supplementary Note 23) Comparing the tact time, which is the expected creation time per unit of the product, with the cycle time, and determining that there is a work delay when the cycle time is longer than the tact time. The program according to Supplementary Note 21 or 22.

[0062] (Supplementary Note 24) Setting the average value or moving average value of the cycle times of a predetermined number of the products as the cycle time. The program according to Supplementary Note 23.

[0063] (Supplementary Note 25) Determining the presence or absence of work delay by comparing the expected output number per unit time of the product with the actual output number per unit time of the product. 25. The program of any one of appendices 21 to 24.

[0064] (Appendix 26) extracting an area specified by a coordinate position in the video as the predetermined area; 26. The program of any one of appendices 21 to 25.

[0065] (Appendix 27) Extracting the predetermined area from the image taken by the imaging device that takes an image of the work site including the predetermined area. 27. The program according to any one of appendices 21 to 26.

[0066] (Appendix 28) processing the images from the image capturing devices provided for each of the work sites in order to capture images of the plurality of work sites including the predetermined area; 27. The program described in Appendix 27. [Explanation of symbols]

[0067] 1 Production management system 2. Control section 3 Recording section 4. Imaging equipment 5 Placement area 6 Network 21 Extraction part 22 Acquisition Department 23 Judgment section 31 Analysis Log 32 Control parameters 51 Workbench

Claims

1. Extraction means for extracting a predetermined area for placing a product that is the result of work by an operator from the video of the work site, Determination means for detecting whether there is an image of a human hand shape in the video of the predetermined area, and recording, as an analysis log, the time when the human hand shape was detected and the time when the human hand shape could no longer be detected after the detection, Management means for adding the actual output of the product when the elapsed time from the time when the human hand shape was detected to the time when the human hand shape could no longer be detected after the detection of the human hand is greater than a predetermined value, A production management system comprising:

2. The management means records the elapsed time from the time when the human hand shape can no longer be detected to the time when the human hand shape is detected next as the cycle time, which is the time for creating the product, The production management system according to claim 1.

3. The management means compares the cycle time, which is the expected creation time per unit of the product, with the cycle time, and determines that there is a work delay when the cycle time is longer than the tact time, The production management system according to claim 2.

4. The management means sets the average value or the moving average value of the cycle times of a predetermined number of the products as the cycle time, The production management system according to claim 3.

5. The management means determines the presence or absence of work delay by comparing the expected output number per unit time of the product with the actual output per unit time, The production management system according to claim 1. 【Claim ⑥】 The extraction means extracts, as the predetermined area, an area specified by the coordinate position in the video, The production management system according to claim 1.

7. Comprising a photographing device for photographing the work site including the predetermined area, The extraction means extracts the predetermined area from the video from the photographing device The production management system according to any one of claims 1 to 6.

8. The photographing device is provided for each work site in order to photograph a plurality of work sites including the predetermined area, The extraction means, the determination means, and the management means each perform processing on the video obtained for each photographing device, The production management system according to claim 7.

9. Extract a predetermined area for placing a product that is the result of work by an operator from the video of the work site, Detect whether there is an image of a human hand shape in the image of the predetermined area, and record the time when the human hand shape is detected and the time when the human hand shape can no longer be detected after detection as an analysis log. When the elapsed time from the time when the human hand shape is detected to the time when the human hand shape can no longer be detected after the detection of the human hand is greater than a predetermined value, add the actual output of the product. A production management method performed by a computer executing this.

10. Extract a predetermined area for placing a product that is the result of the worker from the image of the work site. Detect whether there is an image of a human hand shape in the image of the predetermined area, and record the time when the human hand shape is detected and the time when the human hand shape can no longer be detected after detection as an analysis log. When the elapsed time from the time when the human hand shape is detected to the time when the human hand shape can no longer be detected after the detection of the human hand is greater than a predetermined value, add the actual output of the product. A program for production management that causes a computer to execute this.

Images