JP2018005789A - Work estimation device, work estimation method and work estimation program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a work estimation device, a work estimation method and a work estimation program capable of accurately estimating a series of work elements included in a work.SOLUTION: A work estimation device of a work estimation system comprises: a storage part 320 for storing reference data in which, prescribed quantization processing is performed with respect to time series data of a behavior detected from a worker who performed work elements, in association with each of the work elements including plural operation elements, the work elements being included in the work; a signal processing part for performing the prescribed quantization processing to the time series data detected by a detection part 120 attached to a portable terminal of the worker, when the worker performed the work; and an estimation part for, based on comparison of the reference data stored in the storage part and the data to which the processing is performed by the signal processing part, estimating the work elements which sequentially appear in the time series data.SELECTED DRAWING: Figure 3

Description

  The present invention relates to a work estimation device, a work estimation method, and a work estimation program.

  2. Description of the Related Art Conventionally, techniques for improving the work efficiency of workers in, for example, logistics processes, factories, and offshore are known. In this connection, Patent Document 1 describes a technique for automatically recording a work history of a worker who mainly works in a greenhouse. Specifically, workers who entered and exited were identified based on the entrance and exit times that responded to people by sensors installed at the entrances and exits of multiple greenhouses. Based on this, the work of the worker in the facility area is recorded.

  According to Patent Document 2, in accordance with instructions for starting and ending measurement of work time for each process in a production line including a plurality of processes, the work time of each process is repeatedly measured, and based on the measured work time of each process. It is described that the average work time is calculated and displayed for each process included in the production line.

  In Patent Document 3, in the pharmacy business monitoring system, the work in each work area set in the pharmacy is photographed to capture the worker performing the business and the mark display object handled by the worker. Further, it is described that when a mark in an image is recognized, an index including a shooting time of the mark and a work area ID is created and image data is stored in association with the index.

  In Patent Literature 4, event data is extracted with time by determining whether or not an output value of a predetermined sensor value satisfies a predetermined output condition, and the state of a regular work procedure that is a series of work procedures. It describes that the work content is estimated from the work graph indicated by the transition and the time series information of the actual event data.

  Non-Patent Document 1 traces the movement of a picking worker by collecting geomagnetism or acceleration using a dedicated terminal (clock type) for picking work in a factory and absorbing differences in movement such as wrinkles by the worker. A technique for judging the accuracy of the picking work is described.

  Non-patent document 2 describes numerically how to move the upper body (angle of the upper body) and difficulty (questionnaire results) when taking out the load from the shelf using a smartphone in a breast pocket. The estimation is described.

JP 2013-3925 A JP2013-211000A JP 2010-258729 A JP 2011-184121 A

"Picking monitoring system in the factory using acceleration sensor", 74th National Convention of Information Processing Society of Japan (2012/3), Jun Kono, Nagatsugu Sato, Satoshi Takayama, Yoshitoshi Murata, Iwate Prefectural University "A study on workload recognition in a work probe system in a warehouse using a smartphone", Proceedings of the IEICE General Conference (2016/3), Yuka Sato, Toshimasa Aso, Tokyo Marine University

  However, the techniques described in Patent Documents 1 to 4 and Non-Patent Documents 1 and 2 described above do not estimate each of a series of work elements included in one work.

  Further, the shipping operation includes a series of work elements such as moving with a cart, taking a load, placing a load on the cart, moving with the cart, and unloading the load from the cart. Each of the series of work elements includes, for example, the basic operation elements of the worker such as grabbing, releasing, moving, and stopping the article. These operating elements often vary greatly depending on the actual situation at the work site. Specifically, there is a high possibility that the motion elements greatly differ depending on the place of work, the distance to the shelf on which the article is placed, the height of the shelf, the physique of the worker, and the like. For this reason, when trying to estimate the contents of the motion element from the time-dependent data of acceleration and angle, there is a large error due to individual differences, and it is difficult to accurately estimate the work.

  The present invention has been made in view of such circumstances, and provides a work estimation device, a work estimation method, and a work estimation program capable of accurately estimating a series of work elements included in a work. One of the purposes.

  One aspect of the present invention is a work element included in a work, and is associated with each of the work elements that can include a plurality of motion elements, and time-series data of behavior detected from the worker who performed the work element A storage unit that stores reference data subjected to a predetermined quantization process, and time-series data detected by a detection unit attached to the worker when the worker performs work. A processing element that sequentially appears in the time-series data based on a comparison between the signal processing unit that performs the quantization processing of the reference data and the reference data stored in the storage unit and the data processed by the signal processing unit. And an estimation unit for estimation.

  According to one aspect of the present invention, a series of work elements included in a work can be estimated with high accuracy.

It is a figure which shows an example of the flow of a collection work. It is a lineblock diagram showing an example of work presumption system 1 of a 1st embodiment. 2 is an internal configuration diagram illustrating an example of a mobile terminal 100 and a work estimation device 300. FIG. It is a flowchart which shows an example of the flow of the process which produces | generates reference data. It is a figure which shows an example of behavior data. It is a figure which shows the behavior data of the Y-axis acceleration among behavior data. It is a figure which shows an example of the reference data which converted the behavior data of the Y-axis acceleration by the predetermined | prescribed quantization process. 10 is a flowchart illustrating an example of a flow of processing for estimating a work element and processing for providing information to the management terminal 400. It is a figure which shows an example of a Y-axis acceleration among behavior data. It is a figure which shows an example of the data which performed the quantization process with respect to the Y-axis acceleration among behavior data. It is a figure which shows an example of a comparison process. It is a figure which shows the experimental change which estimated the work element with the time change of the actual work element in the 1st collection work, and the work estimation apparatus. It is a figure which shows the experimental result which estimated the work element with the time change of the actual work element in the 2nd collection work, and the work estimation apparatus. It is a figure which shows an example of the provision information 360A and 360B provided to the management terminal 400. It is a block diagram which shows the modification of the work estimation system 1 of 1st Embodiment. It is a block diagram which shows an example of the work estimation system 1A of 2nd Embodiment. It is a figure which shows an example of the physical distribution management database 612. It is a figure which shows an example of the provision information 360C in 2nd Embodiment. It is a figure which shows an example of the provision information 360D in 2nd Embodiment. It is a block diagram of the work estimation system 1B of a modification.

  Hereinafter, embodiments of a work estimation device, a work estimation method, and a work estimation program according to the present invention will be described with reference to the drawings. The work estimation system to which the present invention is applied is a work element included in a worker's work using a terminal held by the worker in order to improve work efficiency at a work site such as a distribution site, a factory, or offshore. Is a system for estimating each of the above. “Work” is an action repeatedly performed by a worker at a work site, and includes a series of work elements. The “work element” is a series of actions of an operator that can include a plurality of operation elements for completing the work. “Operation element” is a unit action of the worker for the worker to complete each work element, for example, raising a hand, squatting, walking, grabbing a load, or unloading a load. An action that tends to have uniform characteristics in series data.

  Hereinafter, a case where the “work” is a collection work for sorting and shipping physical distribution objects will be described. FIG. 1 is a diagram illustrating an example of a flow of a collection work. For example, the worker P1 searches for a load 10 (an example of a physical distribution object) from a shelf 12 installed in a warehouse, takes out the searched load 10, and places the extracted load 10 on a carriage 14. Next, the worker P1 moves the carriage 14 on which the load 10 is loaded to the collection place, and places the load 10 on the vehicle 20 at the collection place. Next, the worker P1 moves the carriage 14 on which the load 10 is not placed toward the shelf 12 where the next load 10 is located.

  In the above-described collection work, the “work” includes three types of “work elements” that move toward the load together with the cart, and transport the load with the cart, search for the load, and pick up the load. That is, the three types of work elements are “move”, “search”, and “take out”. The work element “movement” includes, for example, an “operation element” of grabbing a cart or walking. The work element “search” includes, for example, “motion elements” such as moving the head left and right, twisting the body, and walking. The work element “removal” includes “motion elements” such as reaching for the load, grasping the load, and unloading the load.

[First Embodiment]
(Configuration of work estimation system 1)
The first embodiment will be described below. FIG. 2 is a configuration diagram illustrating an example of the work estimation system 1 according to the first embodiment. The work estimation system 1 includes, for example, a mobile terminal 100, a work estimation apparatus 300, and a management terminal 400. The portable terminal 100, the management terminal 400, and the work estimation apparatus 300 are connected via a network (not shown). The network includes, for example, a wireless base station, a Wi-Fi access point, a communication line, a provider, the Internet, and the like. Note that not all combinations of these components need to be able to communicate with each other, and the network may partially include a local network.

  The portable terminal 100 is a portable terminal device such as a smartphone or a mobile phone that is carried by the worker P1. The portable terminal 100 is accommodated in the breast pocket of the worker P1, for example. The portable terminal 100 transmits behavior data related to the behavior of the worker P1 to the work estimation device 300. In the present embodiment, the device that transmits the behavior data is the mobile terminal 100, but is not limited thereto, and may be a computer built in an attachment such as a helmet or a watch.

  The work estimation device 300 is a server device that performs processing on behavior data. The work estimation device 300 transmits information related to the work element estimation result to the management terminal 400 as a processing result. The management terminal 400 is a terminal device such as a personal computer operated by a warehouse or luggage manager P2. The management terminal 400 presents information to the manager P2 based on information related to the work element estimation result.

  FIG. 3 is an internal configuration diagram illustrating an example of the mobile terminal 100 and the work estimation device 300. The mobile terminal 100 includes, for example, a terminal side communication unit 110, a sensor unit 120, a user interface unit 130, and a control unit 140.

  The terminal-side communication unit 110 is a communication interface such as a NIC (Network Interface Card) or a wireless communication module. The sensor unit 120 includes, for example, an acceleration sensor 122 and an angle sensor 124. The acceleration sensor 122 detects, for example, the X-axis and Y-axis accelerations in the planar direction and the Z-direction acceleration in the vertical direction. The angle sensor 124 detects a pitch angle, a roll angle, and a yaw angle. The sensor unit 120 may acquire position information (latitude and longitude) using GNSS (Global Navigation Satellite System). The user interface unit 130 includes various devices such as a touch panel that receives an operation of the worker P1 and displays an image, and a speaker.

  The control unit 140 is realized by, for example, a processor such as a CPU (Central Processing Unit) executing a program stored in a program memory. Also, part or all of the control unit 140 may be realized by hardware such as LSI (Large Scale Integration), ASIC (Application Specific Integrated Circuit), or FPGA (Field-Programmable Gate Array), or software. And hardware may cooperate. The control unit 140 uses the terminal-side communication unit 110 to obtain behavior data including acceleration data in the X-axis, Y-axis, and Z-direction, and angle data of pitch angle, roll angle, and yaw angle. Send to. The behavior data corresponds to an example of time-series data of behavior detected from the worker P1 who performed the work element.

  The work estimation apparatus 300 includes, for example, a communication unit 310, a reference data storage unit 320, a reference data generation unit 330, a signal processing unit 340, an estimation unit 350, and a provision information generation unit 360. The communication unit 310 is a communication interface such as a NIC or a wireless communication module.

  The reference data storage unit 320 uses, for example, an HDD (Hard Disc Drive), a flash memory, an EEPROM (Electrically Erasable Programmable Read Only Memory), a ROM (Read Only Memory), a RAM (Random Access Memory), or a plurality of these. This is realized by a hybrid storage device. The storage unit stores various programs such as firmware and application programs, processing results by various functional units, and the like. Part or all of the reference data storage unit 320 may be realized by an external storage device that can be accessed by the work estimation device 300 via various networks. An example of the external storage device is a NAS (Network Attached Storage) device. The reference data storage unit 320 stores reference data that is data for estimating the work element of the worker P1. Further, the reference data storage unit 320 may store other information such as a processing result in the work estimation device 300.

  The reference data generation unit 330, the signal processing unit 340, the estimation unit 350, and the provision information generation unit 360 are realized by, for example, a processor such as a CPU executing a program stored in a program memory. Some or all of these functional units may be realized by hardware such as LSI, ASIC, or FPGA, or may be realized by cooperation of software and hardware.

  The reference data generation unit 330 generates reference data. The signal processing unit 340 performs a predetermined quantization process on the behavior data received by the communication unit 310. The estimation unit 350 estimates work elements that sequentially appear in the behavior data based on a comparison between the reference data and the data that has been subjected to the predetermined quantization processing by the signal processing unit 340. The provision information generation unit 360 generates provision information to be provided to the administrator P2 based on the estimation result of the estimation unit 350.

(Reference data generation process)
FIG. 4 is a flowchart illustrating an example of a flow of processing for generating reference data. The process of generating the reference data is executed for each work element registered in the work estimation apparatus 300. Specifically, the process of generating the reference data is executed for each of the three types of “work elements” of “move”, “search”, and “take out”.

  First, the portable terminal 100 determines whether or not the worker P1 is stationary based on the acceleration and direction detected by the acceleration sensor 122 and the angle sensor 124 (step S100). When the worker P1 is stationary, the portable terminal 100 outputs a voice announcement (step S102). Using the user interface unit 130, the mobile terminal 100 outputs, for example, a voice “Please say the work element to be registered.” The portable terminal 100 starts a voice recognition process (step S104).

  The portable terminal 100 determines whether or not the character string recognized by the voice recognition process is a work element (step S106). If the character string is a work element, the portable terminal 100 acquires behavior data (step S108). The portable terminal 100 determines whether or not the acquisition of behavior data has been completed (step S110). For example, the portable terminal 100 determines that the acquisition of the behavior data is completed when the voice “end” is acquired by the voice recognition process. In addition, the portable terminal 100 complete | finishes a process, when an audio | voice is not a work element in step S106.

  Next, the work estimation apparatus 300 takes in behavior data acquired by the mobile terminal 100 (step S112). The work estimation device 300 receives behavior data by receiving the terminal-side communication unit 110 transmitted from the mobile terminal 100 using the communication unit 310. FIG. 5 is a diagram illustrating an example of behavior data. FIG. 6 is a diagram illustrating behavior data of Y-axis acceleration among the behavior data. For example, voice recognition processing is performed at times t1 and t4, the start time and end time of behavior data acquisition of Y-axis acceleration are recorded, and acquisition of behavior data is detected. In addition, the work estimation device 300 extracts the necessary behavior data for the period of time t2 to t3 from the acquired behavior data by detecting the stationary state of the worker P1 at times t1 to t2 and t3 to t4. It can be recorded in the reference data storage unit 320.

  Next, the work estimation apparatus 300 performs a predetermined quantization process by the reference data generation unit 330 (step S114). The predetermined quantization process is, for example, a process of leveling acceleration data of the X-axis, Y-axis, and Z-axis acceleration, and data of the pitch angle, roll angle, and yaw angle. The leveling process is, for example, a process of classifying a value included in a predetermined range as “1”, a value exceeding a predetermined range as “2”, and a value less than the predetermined range as “0”. The work estimation apparatus 300 converts, for example, behavior data into data that changes between three values of 0, 1, and 2 by a predetermined quantization process. FIG. 7 is a diagram illustrating an example of reference data obtained by converting the behavior data of the Y-axis acceleration by a predetermined quantization process. The reference data shown in FIG. 7 is data quantized in the period from time t2 to t3 when the behavior data is acquired.

  Next, the work estimation apparatus 300 stores the data (reference data) that has been subjected to the predetermined quantization processing in the reference data storage unit 320 (step S116).

  The work estimation device 300 may customize the reference data for each type of the load 10 that performs the collection work, the type of the shelf 12 or the carriage 14, and the warehouse, for example, and the type of the load 10, the shelf 12 or the carriage 14 The type and warehouse may be specified so that customized reference data can be selected.

(Work element estimation processing and information provision processing)
FIG. 8 is a flowchart illustrating an example of the flow of processing for estimating work elements and processing for providing information to the management terminal 400. The process of estimating the work element and the process of providing information to the management terminal 400 are repeatedly executed every predetermined period. However, the present invention is not limited to this, and may be started at the timing when the behavior data is acquired.

  First, the work estimation apparatus 300 acquires time-series behavior data from the mobile terminal 100 using the communication unit 310 (step S200). Next, the work estimation apparatus 300 uses the signal processing unit 340 to perform a predetermined quantization process on the acquired behavior data (step S202). The predetermined quantization process in the signal processing unit 340 is the same process as the predetermined quantization process in the reference data generation unit 330.

  Next, the work estimation apparatus 300 selects one reference data stored in the reference data storage unit 320 (step S204). Next, the work estimation apparatus 300 uses the estimation unit 350 to compare the quantized data with the reference data (step S206). The work estimation apparatus 300 uses a technique called DP matching, for example, to expand or compress the time axis of the reference data and compare both data. In addition, the work estimation apparatus 300 may compress or expand the time axis of the data subjected to the quantization process, and may expand or compress both time axes. Thereby, the work estimation apparatus 300 can compare both data by reducing the difference between the reference data and the time required for the actual collection work.

  Next, the work estimation apparatus 300 calculates a score as a result of the comparison (step S208). For example, the work estimation device 300 increases the score when the values of both data compared at a certain time match, and the values of both data compared at a certain time do not match. If so, lower the score. Note that the work estimation apparatus 300 may calculate a cost that represents a rate at which both data do not match, as well as a score, and further calculate a similarity that represents a rate at which both data match. May be.

  Next, the work estimation apparatus 300 determines whether or not all reference data has been selected for the data subjected to the predetermined quantization process (step S210). If the work estimation apparatus 300 has not selected all the reference data, the work estimation apparatus 300 returns the process to step S204.

  When all the reference data is selected, the work estimation device 300 estimates that the work element with the highest score is a work element that has appeared in the acquired behavior data (step S212).

  Next, the work estimation device 300 estimates the entire work based on the estimated work elements (step S214). The work estimation device 300 estimates that the whole work is a collection work when three kinds of work elements such as movement, search, and take-out appear sequentially.

  Next, the work estimation apparatus 300 determines whether or not the transmission timing of the estimation result has arrived (step S216). For example, when receiving a command requesting an estimation result from the management terminal 400, the work estimation device 300 determines that the transmission timing of the estimation result has arrived. The work estimation device 300 may determine that the transmission timing of the estimation result has arrived in response to the arrival of a predetermined repetition period, not limited to the timing of receiving the command.

  When the transmission timing of the estimation result has arrived, the work estimation device 300 generates the provision information by the provision information generation unit 360 (step S218). Next, the work estimation device 300 transmits the generated provision information to the management terminal 400 (step S220).

(Work element estimation process)
FIG. 9 is a diagram illustrating an example of the Y-axis acceleration in the behavior data. It is assumed that the work estimation apparatus 300 acquires behavior data as shown in FIG. 9 in step S200 described above. FIG. 10 is a diagram illustrating an example of data obtained by performing quantization processing on the Y-axis acceleration among the behavior data. The work estimation apparatus 300 obtains data as shown in FIG. 10 by cutting out a period of time between substantially stationary periods and performing a predetermined quantization process on the behavior data shown in FIG. .

  FIG. 11 is a diagram illustrating an example of the comparison process. The estimation unit 350 includes the data D1-1 included in the frame of the predetermined period among the data D1 subjected to the quantization process illustrated in FIG. 11A, the reference data D11 illustrated in FIG. The reference data D12 shown in c) and the reference data D13 shown in FIG. The estimation unit 350 converts the data D1-2 shifted from the data D1-1 by the offset period, the reference data D11 shown in FIG. 11B, the reference data D12 shown in FIG. 11C, and FIG. Each of the indicated reference data D13 is compared. Furthermore, the estimation unit 350 performs the data D1-3 shifted from the data D1-2 by the offset period, the reference data D11 shown in FIG. 11B, the reference data D12 shown in FIG. 11C, and FIG. And the reference data D13 shown in FIG. In this way, the estimation unit 350 repeats the process of comparing the data shifted by the offset period with each of the reference data.

  When the estimation unit 350 compares the data included in the frame of the predetermined period with the reference data, the estimation unit 350 performs a process (scaling process) to compress or expand the reference data on the time axis, and the reference data subjected to the scaling process And the data included in the frame of the predetermined period are compared. For example, the estimation unit 350 compares the reference data after performing scaling processing such as 1.1 times, 1.2 times, 0.9 times, and 0.8 times on the time axis. In that case, the estimation unit 350 may use the magnification applied to the reference data corresponding to one work element for the reference data corresponding to another work element.

  Then, the estimation unit 350 calculates a score for each data included in the frame for the predetermined period as a comparison result, and estimates a work element corresponding to the highest score for each data included in the frame for the predetermined period. Thereby, the estimation unit 350 can estimate the work elements in the order of appearance of the work elements included in the work.

  In addition, when work elements appear continuously, the estimation unit 350 can estimate the required time from the start time to the end time of each work element by adding the work elements.

(Overall work estimation process)
FIG. 12 is a diagram showing temporal changes in actual work elements in the first collection work, and experimental results of estimating work elements by the work estimation device 300. The true value in FIG. 12 represents a change in work element corresponding to the actual work of the worker. The work element ID is information for specifying a work element.

  In the first collection work, the vehicle moves toward the load 10 together with the carriage 14 (ID = 1), searches for the load 10 (ID = 2), takes out the load 10 (ID = 0), and the carriage on which the load 10 is loaded. Work elements appear in the order of movement (ID = 1). It is understood that the work estimation device 300 can estimate the work elements in the order of 1, 2, 0, 1 by estimating the work elements that appear sequentially. Also, the change in work element ID shown in FIG. 12 (a) and the change in work element ID shown in FIG. 12 (b) are different for workers, and therefore the time required for each work element is different. Even in this case, according to the work estimation device 300, it can be seen that the work element estimation result close to the true value can be obtained.

  The work estimation device 300 stores information representing a time-series pattern of work element IDs corresponding to the true value of the first collection work, for example, in the reference data storage unit 320. The estimation unit 350 determines from the degree of similarity whether the array of work element IDs corresponding to the estimated work elements is in the order of 1, 2, 0, 1. In addition, the estimation part 350 should just compare the estimated result and the order set beforehand, and should just calculate a similarity (or cost) similarly to the comparison process mentioned above. When the arrangement of the work element IDs corresponding to the estimated work elements is similar to the preset order of 1, 2, 0, 1, the estimation unit 350 performs the work performed by the worker P1 as the first collection work. Determine that there is. When the arrangement of the work element IDs corresponding to the estimated work elements is not similar to the preset order of 1, 2, 0, 1, the estimation unit 350 determines that the work performed by the worker P1 is the first collection work. Judge that there is no. Thereby, the work estimation apparatus 300 can estimate the entire work.

  FIG. 13 is a diagram illustrating a temporal change of an actual work element and a result of an experiment in which the work element is estimated by the work estimation device 300 in the second collection work. The second collection work is performed in the order of moving toward the load 10 together with the carriage 14 (ID = 1), taking out the load 10 (ID = 0), and moving the load 10 together with the carriage 14 (ID = 1). An element appears. It is understood that the work estimation device 300 can estimate the work elements in the order of 1, 0, 1 by estimating the work elements that appear sequentially. Further, the change in the work element ID shown in FIG. 13A and the change in the work element ID shown in FIG. 13B are different for the workers, and therefore the time required for each work element is different. Even in this case, according to the work estimation device 300, it can be seen that the work element estimation result close to the true value can be obtained.

  The work estimation device 300 stores information representing a time-series pattern of work element IDs corresponding to the true value of the first collection work, for example, in the reference data storage unit 320. The estimation unit 350 determines from the degree of similarity whether the array of work element IDs corresponding to the estimated work elements is in the order of 1, 0, or 1. When the arrangement of the work element IDs corresponding to the estimated work elements is similar to the preset order of 1, 0, 1, the estimation unit 350 indicates that the work performed by the worker P1 is the second collection work. Determine. The estimation unit 350 determines that the work performed by the worker P1 is not the second collection work when the arrangement of the work element IDs corresponding to the estimated work elements is not similar to the preset order of 1, 0, 1 Determine. Thereby, the work estimation apparatus 300 can estimate the entire work.

(Providing information generation process)
FIG. 14 is a diagram illustrating an example of provision information 360 </ b> A and 360 </ b> B provided to the management terminal 400. The provided information 360A and 360B is information for displaying an image in which the ratio of the required time for each work element estimated by the estimation unit 350 is graphed. The provided information generation unit 360 calculates the ratio of the required time for each work element estimated by the estimation unit 350 in the work time of the worker P1. The provided information generation unit 360 generates information for displaying a pie chart based on the calculated ratio of the required time for each work element.

  The work estimation apparatus 300 transmits the provision information 360A to the management terminal 400 at the transmission timing of the estimation result, and displays the image graphed by the management terminal 400. Thereby, the work estimation apparatus 300 can display a pie chart corresponding to the provision information 360A. As a result, the manager P2 can intuitively recognize that the search rate in the collection work is large, and can perform improvement work such as the arrangement of the load 10 and the shelf 12.

  In response to the arrival of the estimation result transmission timing after the arrangement of the load 10 or the shelf 12 has arrived, the work estimation device 300 transmits the provision information 360B to the management terminal 400, and displays the image graphed by the management terminal 400. Display. Thereby, the work estimation apparatus 300 can display a pie chart corresponding to the provision information 360B. As a result, the administrator P2 can intuitively recognize that the search rate has decreased, and can recognize that improvement work such as the placement of the load 10 and the shelf 12 has been effectively performed. it can.

(Attitude extraction process)
The provided information generation process may include a process of extracting a posture with a heavy burden on the worker P1 from postures during the work. FIG. 15 is a configuration diagram illustrating a modification of the work estimation system 1 according to the first embodiment. In this case, the work estimation device 300 further includes a camera device 200 that images the space in the warehouse, acquires moving image data as a result of imaging, and sends the moving image data to the communication unit 310. From the moving image data acquired by the communication unit 310, a posture with a heavy burden is extracted from the posture of the worker P1. The posture with a large load is, for example, a posture in which the worker P1 bends his / her waist greatly in order to take the load 10 on the floor. The work estimation device 300 identifies the work element estimated at the time of the moving image data from which the posture with a large burden is extracted, and estimates that the load on the worker P1 in the identified work element is high.

(Effects of the first embodiment)
According to the work estimation system 1 of the first embodiment, reference data obtained by performing predetermined quantization processing on time-series data of behavior detected by an operator is stored, and reference data and behavior data are stored. By estimating the work elements that appear sequentially in the behavior data based on the comparison with the data subjected to the quantization process, a series of work elements included in the work can be estimated with high accuracy. As described above, the motion element has uniform characteristics that appear in the behavior detected from the behavior of the worker P1, and also varies depending on the physique of the worker, the conditions of the work place, etc., so this is estimated by itself. To do this, complex and sophisticated information processing is required. On the other hand, since the detected features differ greatly between the motion elements, the change in the features when viewed as the entire motion element does not depend much on individual differences or differences in conditions, and is easy to discriminate. That is, in the work element, since the difference in the features of each operation element appears in order, it is relatively easy to generalize and discriminate by a computer.

  Further, according to the work estimation system 1 of the first embodiment, the work is performed based on a comparison between the information indicating the appearance order of the plurality of work elements associated with the work and the estimated order of appearance of the plurality of work elements. Can be estimated. Thereby, according to the work estimation system 1, it is possible to monitor whether or not the worker P1 is performing a preset work.

  Furthermore, according to the work estimation system 1 of the first embodiment, it is possible to estimate the time required from the start time to the end time for each work element that appears sequentially. Improvements such as changing the arrangement of 12 can be performed. For example, the arrangement of the loads 10 and the shelves 12 can be changed so as to reduce the ratio of the time required for “searching” work elements.

  Furthermore, according to the work estimation system 1 of the first embodiment, by generating information for displaying an image in which the ratio of the required time for each estimated work element is graphed, the information is graphed to the administrator P2. Images can be presented. Thereby, according to the work estimation system 1, work elements can be recognized in an easy-to-understand manner, and convenience can be improved.

  Furthermore, according to the work estimation system 1 of the first embodiment, the worker P1 in the estimated work element based on the temporal relationship between the posture of the worker P1 extracted from the moving image data and the estimated work element. Therefore, an improvement that suppresses the load on the worker P1 can be performed.

[Second Embodiment]
Hereinafter, the second embodiment will be described. FIG. 16 is a configuration diagram illustrating an example of a work estimation system 1A according to the second embodiment. The work estimation system 1A of the second embodiment displays an image in which the ratio of the required time for each work element estimated by the estimation unit 350 is graphed for each load 10, each shelf 12, or each warehouse. This is different from the first embodiment in that the information is generated. The shelf 12 or the warehouse is an example of a group in which the loads 10 are grouped. Hereinafter, this difference will be mainly described.

  The work estimation system 1A of the second embodiment includes, for example, a physical distribution management server 600 and a handy terminal 500 in addition to the work estimation system 1 of the first embodiment. The physical distribution management server 600 is a server device that is realized when a processor such as a CPU executes a program stored in a program memory. The logistics management server 600 registers information in the logistics management database storage unit 610 for managing the cargo 10. FIG. 17 is a diagram illustrating an example of the distribution management database 612. The logistics management database 612 stores, for example, a warehouse ID, a shelf ID, and a collection time in association with the shipment ID.

  The handy terminal 500 is a terminal device including a processing unit such as a code reading device, a wireless communication NIC, and a CPU, for example. When the worker P1 collects the load 10, the handy terminal 500 acquires the load ID by, for example, reading a code attached to the load 10 with a code reader. The handy terminal 500 adds the time information as the collection time and the worker ID to the ID of the load 10 by the NIC, and transmits it to the physical distribution management server 600 and the work estimation device 300A.

  When the work estimation device 300A receives the load ID, the user ID, and the pickup time from the handy terminal 500, the work estimation device 300A stores the load ID, the user ID, and the pickup time. The work estimation device 300A acquires the work element of the worker P1 corresponding to the worker ID based on the received worker ID. For example, the work estimation device 300A stores in advance a work element estimated based on behavior data detected from the worker P1 in association with the worker ID associated with the identification information of the mobile terminal 100. Thus, the work element of the worker P1 corresponding to the worker ID received from the handy terminal 500 is acquired. In addition, the work estimation device 300A extracts a work element corresponding to the collection time received from the handy terminal 500 from the obtained work element of the worker P1, and obtains a work including the extracted work element. Thereby, the work estimation device 300A can calculate the required time for each work element and the ratio of the required time for the collection work of the load 10 corresponding to the load ID.

  Further, the work estimation device 300A may acquire a warehouse ID or a shelf ID associated with the load ID from the physical distribution management server 600. The work estimation device 300A can calculate the required time for each work element and the ratio of the required time for each warehouse ID associated with the load ID. Moreover, the work estimation device 300A can calculate the required time for each work element and the ratio of the required time for each shelf ID associated with the load ID.

  FIG. 18 is a diagram illustrating an example of provision information 360C according to the second embodiment. The work estimation device 300A can create the provision information 360C in which the relationship between the load ID, the shelf ID, or the warehouse ID and the travel time, the search time, or the take-out time is a bar graph, and can provide it to the management terminal 400. . Thereby, the manager P2 can recognize the load 10, the shelf 12, or the warehouse that requires a long work time.

  FIG. 19 is a diagram illustrating an example of the provision information 360D in the second embodiment. The work estimation device 300 </ b> A can create provision information 360 </ b> C that represents a ratio of required time between work elements for each load 10, each warehouse, or each shelf, and can provide the management information 400 to the management information 400. Thereby, the manager P2 can recognize which work element requires a long work time among the travel time, the search time, and the take-out time.

[Modification]
Hereinafter, modifications of the first embodiment and the second embodiment will be described. FIG. 20 is a configuration diagram of a work estimation system 1B according to a modification. The work estimation system 1B of the modified example is a relay device 700-1, which collects behavior data from the portable terminal 100 of the worker who performs the collection work in the warehouse, for each warehouse W-1,. 700-N. Relay devices 700-1,... 700-N acquire and store behavior data from a plurality of portable terminals 100 using wireless communication or wired communication. The relay device 700-1,..., 700-N, for example, does not collect behavior data during the collection work, and from the portable terminal 100 by a wired cable during the closing time when the collection work is not being performed. Collect behavior data. Relay device 700-1,... 700-N transmits behavior data collected from a plurality of portable terminals 100 to work estimation device 300B, for example, once a day. For example, the work estimation device 300B calculates the ratio of the time required for each work element for the collection work of the load 10 corresponding to the load ID for each worker, for each load 10, for each shelf 12, or for each warehouse W. Can do.

  As mentioned above, although the form for implementing this invention was demonstrated using embodiment, this invention is not limited to such embodiment at all, In the range which does not deviate from the summary of this invention, various deformation | transformation and substitution Can be added.

DESCRIPTION OF SYMBOLS 1, 1A, 1B ... Work estimation system, 10 ... Load, 12 ... Shelf, 14 ... Dolly, 100 ... Portable terminal, 110 ... Terminal side communication part, 120 ... Sensor part, 122 ... Acceleration sensor, 124 ... Angle sensor, 130 ... User interface unit, 140 ... Control unit, 200 ... Camera device, 300, 300A, 300B ... Work estimation device, 310 ... Communication unit, 320 ... Reference data storage unit, 330 ... Reference data generation unit, 340 ... Signal processing unit, 350 ... Estimation unit, 360 ... Provision information generation unit, 360A, 360B, 360C ... Provision information, 400 ... Management terminal, 500 ... Handy terminal, 600 ... Distribution management server, 610 ... Distribution management database storage unit, 612 ... Distribution management database , 700-1, 700-N ... Relay device

Claims (12)

Predetermined quantization on time-series data of behavior detected from the worker who performed the work element in association with each work element that is included in the work and can include a plurality of motion elements A storage unit for storing the processed reference data;
A signal processing unit that performs the predetermined quantization processing on the time-series data detected by the detection unit attached to the worker when the worker performs the work;
Based on a comparison between the reference data stored in the storage unit and the data processed by the signal processing unit, an estimation unit that estimates work elements that appear sequentially in the time series data;
A work estimation device comprising: The storage unit stores information indicating an appearance order of a plurality of work elements in association with the work,
The estimation unit estimates the work based on a comparison between the estimated appearance order of the plurality of work elements and the information.
The work estimation apparatus according to claim 1. The estimation unit estimates a required time from a start time to an end time for each of the work elements.
The work estimation apparatus according to claim 1 or 2. An information generation unit that generates information for displaying an image in which the ratio of the required time for each work element estimated by the estimation unit is graphed;
The work estimation device according to any one of claims 1 to 3. An image acquisition unit that acquires an image of the worker;
The estimation unit estimates a worker's load on the estimated work element based on a temporal relationship between the posture of the worker included in the image acquired by the image acquisition unit and the estimated work element.
The work estimation apparatus according to any one of claims 1 to 4. After the voice received by the portable terminal carried by the worker in a state where the worker is stationary, the work element recognized based on the voice received by the portable terminal, and after the voice is received by the portable terminal A reference data generation unit configured to generate reference data associated with each of the work elements based on time-series data of behavior detected from the worker who performed the work elements;
The work estimation apparatus according to any one of claims 1 to 5. The work elements include movements for taking out and transporting physical distribution objects, searching for physical distribution objects, and taking out physical distribution objects.
The work estimation apparatus according to any one of claims 1 to 6. For each of the logistics objects or the grouping of the logistics objects, the movement for taking out and transporting the logistics objects estimated by the estimation unit, searching for the logistics objects, and requirements for taking out the logistics objects An information generation unit that generates information for displaying an image in which a time or a ratio of the required time is graphed;
The work estimation apparatus according to claim 7. Predetermined quantization on time-series data of behavior detected from the worker who performed the work element in association with each work element that is included in the work and can include a plurality of motion elements Store the processed reference data,
Applying the predetermined quantization process to the time-series data detected by the detection unit attached to the worker when the worker performs the work,
Based on a comparison between the reference data and the data subjected to the quantization process, a work element that sequentially appears in the time series data is estimated.
Work estimation method. When storing the reference data,
Accept voice from a mobile device carried by the worker while the worker is stationary,
Recognizing work elements based on the voice received from the mobile terminal;
Detecting time-series data of behavior from the worker who performed the work element after the voice was received by the mobile terminal;
Applying the predetermined quantization process to the detected time series data,
Storing the data subjected to the predetermined quantization processing in the recognized work element;
The work estimation method according to claim 9. On the computer,
Predetermined quantization on time-series data of behavior detected from the worker who performed the work element in association with each work element that is included in the work and can include a plurality of motion elements Store the processed reference data in the storage unit,
When the worker performs work, the predetermined quantization processing is performed on the time series data detected by the detection unit attached to the worker,
Based on the comparison between the reference data and the quantized data, work elements that appear sequentially in the time series data are estimated.
Work estimation program. When storing the reference data,
On the computer,
Receive voice from the mobile device carried by the worker while the worker is stationary,
Recognize work elements based on the voice received from the mobile terminal,
After the voice is received by the portable terminal, the time series data of the behavior is detected from the worker who performed the work element,
The predetermined time-series data is subjected to the predetermined quantization process,
Storing the data subjected to the predetermined quantization processing in the recognized work element;
The work estimation program according to claim 11.

Images