WO2019087275A1 - Work analysis device and work analysis method - Google Patents

Abstract

The purpose of the present invention is to perform a series of processes, from determination of a suitable management granularity through to extraction of improvements to be made, without relying on human intervention. This work analysis device is characterized by being provided with: a management granularity determination unit which determines a management granularity for each operation on the basis of work result information obtained by accumulating information relating to a plurality of performed pieces of work; an evaluation value calculation unit which calculates an evaluation value for each of a plurality of evaluation indices for each piece of work on the basis of the work result information; a work classification unit which groups the plurality of pieces of work into management granularity groups in accordance with the determined management granularity for each operation, and classifies an excellent piece of work from among a plurality of pieces of work belonging to each management granularity group on the basis of the calculated evaluation value for each of the plurality of evaluation indices for each piece of work; and a work improvement extraction unit which extracts improvements to be made to each non-excellent piece of work belonging to each management granularity group, on the basis of the evaluation values for the excellent piece of work belonging to the management granularity group.

Description

Work analysis apparatus and work analysis method

The present invention relates to a work analysis device and a work analysis method.

In order to improve the production efficiency of a product, it is necessary to analyze work time and extract improvement points of work. To analyze work time, work needs to be grouped by appropriate control granularity.

For example, when analyzing the manufacturing operation of non-mass-produced products such as elevators, the manufacturing particle size and manufacturing method differ depending on the product specification, so the control particle size is determined by combining the work content in each process, the product specification, and the worker Is desirable.

As a technique for determining the management granularity, for example, in Patent Document 1, “Work type classification unit 11 that generates work time history data by classifying for each work type item value included in work time history data, and work time history data Representative reference time calculation unit 12 that calculates representative reference time for each work type item value, attribute item classification unit 13 that generates work time history data by classifying work time history data for each attribute item value, and work time history data Subdivision reference time calculation unit 14 that calculates the subdivision reference time for each attribute item value from data, and the division reference time determined to be significant are set as reference time of the attribute item value, and for attribute item values that are not so Has a significance evaluation unit 15 for setting a representative reference time as a reference time, and a reference time estimation apparatus is described.

JP, 2015-148961, A

In the technology described in Patent Document 1, the management particle size is calculated by determining whether the difference in work time of each attribute is significant. However, when the number of data of work results is small, it is determined whether it is significant. I can not

Also, conventionally, there is a technology from determination of control particle size to extraction of improvement points, but the evaluation index of work is only work time, and in order to use other evaluation indices, visual confirmation by human Was necessary.

The present invention has been made in view of such circumstances, and it is an object of the present invention to enable a series of processes from determination of appropriate control granularity to extraction of improvement points to be performed without relying on human power. I assume.

Although this application contains multiple means to solve at least one part of the said subject, if the example is given, it is as follows. In order to solve the above problems, a work analysis apparatus according to an aspect of the present invention determines a management particle size determination unit that determines a management particle size for each process based on work performance information in which information on a plurality of executed work is accumulated. The evaluation value calculation unit calculates an evaluation value of each of a plurality of evaluation indices for each work based on the work performance information, and the work is grouped into a management particle size group according to the determined management granularity for each process, A work classification unit that classifies excellent work from the plurality of works belonging to each management granularity group based on the evaluation value of each of the plurality of evaluation indices calculated, and the evaluation value of the excellent work of each management granularity group And a work improvement point extraction unit for extracting a work improvement point of a non-excellent work belonging to each of the control granularity groups.

According to the present invention, it is possible to execute a series of processes from determination of appropriate control granularity to extraction of improvement points without relying on human power. Problems, configurations, and effects other than those described above will be apparent from the description of the embodiments below.

It is a block diagram showing an example of composition of a work analysis system which is one embodiment concerning the present invention. It is a figure which shows an example of the data structure of work performance information. It is a figure which shows an example of the data structure of management particle size information. It is a figure which shows an example of the data structure of flow line information. It is a figure which shows an example of the data structure of evaluation value information. It is a figure which shows an example of the data structure of work classification information. It is a figure which shows an example of the data structure of evaluation value information classified by area. It is a figure which shows an example of the data structure of work improvement point information. It is a figure which shows an example of the data structure of work improvement point library information. It is a flow chart explaining an example of work analysis processing by a work analysis system. It is a figure which shows an example which visualized the dispersion condition of the working time of each analysis particle size. It is a figure which shows an example of the calculation result of the average working time of each analysis particle size. It is a figure which shows an example of the calculation result of the absolute value error of each analysis particle size. It is a figure which shows an example of an output screen. It is a figure which shows an example of an edit screen. It is a flow chart explaining an example of edit processing. It is a block diagram showing an example of composition of a computer.

Hereinafter, an embodiment according to the present invention will be described based on the drawings. In all the drawings for describing the embodiment, the same reference numeral is attached to the same member in principle, and the repetitive description thereof will be omitted. Further, in the following embodiments, the constituent elements (including element steps and the like) are not necessarily essential except in the case where they are particularly clearly shown and where they are considered to be obviously essential in principle. Needless to say. In addition, when we say “consists of A”, “consists of A”, “have A”, and “include A”, except for those cases where it is clearly stated that it is only that element, etc., the other elements are excluded. It goes without saying that it is not something to do. Similarly, in the following embodiments, when referring to the shapes, positional relationships and the like of components etc., the shapes thereof are substantially the same unless particularly clearly stated and where it is apparently clearly not so in principle. It is assumed that it includes things that are similar or similar to etc.

In the present specification, "operation" refers to a plurality of delimiting units that form a series of operations for manufacturing a product. The work is assumed to have information indicating the process, work content, model, and worker as the attribute information.

Also, “granularity” represents a combination of conditions when grouping work into groups in order to analyze work by process. For example, since a series of operations for manufacturing an elevator consists of a plurality of steps, there are different work contents in the same step, different types of machines to be manufactured, and workers are different. It is possible to group into groups based on combinations of processes, work contents, models, and workers.

Specifically, although the control granularity is only a process, it may be a combination of a process and at least one of a model, work content, and a worker.

For example, when grouping each operation under the least condition, the control granularity may be only the process. Moreover, "grain size is coarse" means that there are few conditions when grouping. On the contrary, "fine-grained" means that there are many conditions for grouping.

<Configuration Example of Work Analysis System According to One Embodiment of the Present Invention>
FIG. 1 is a block diagram showing an example of the configuration of a work analysis system according to an embodiment (hereinafter referred to as the present embodiment) according to the present invention.

The work analysis system 100 is configured by connecting a work analysis apparatus 101 to a user terminal 103 and a database 104 via a network 102.

The work analysis apparatus 101 analyzes work that has already been performed for each process to extract work improvement points. Here, the work improvement point refers to a point to be improved, for example, a long work time is required in the work that has already been performed.

The network 102 is a two-way communication network including, for example, a Local Area Network (LAN), a Wide Area Network (WAN), a Virtual Private Network (VPN), the Internet, and the like.

The user terminal 103 comprises, for example, a personal computer, etc., accepts a user's operation of designating a process to be analyzed, newly registering a record of the work improvement point library information 128, editing it, and so forth. , And transmitted to the work analysis apparatus 101 via the network 102. Further, the user terminal 103 displays an output screen 1300 (FIG. 14) or the like representing an analysis result and the like supplied from the work analysis apparatus 101 on a display (not shown) and presents the same to the user.

The database 104 stores, for example, a system such as a MES (Manufacturing Execution System) or data based thereon. Specifically, the work record information in which the information about the executed work is stored is stored, and among the stored work record information, the work record information on the process designated as the process to be analyzed is analyzed The device 101 is supplied.

The work analysis device 101 will be described in detail. The work analysis apparatus 101 includes, for example, a personal computer or a server computer, and includes an arithmetic unit 110, a storage unit 120, an input unit 130, and an output unit 140.

The calculation unit 110 includes a management particle size determination unit 111, an evaluation value calculation unit 112, a work classification unit 113, and a work improvement point extraction unit 114.

Based on the work record information 121 (details described later) supplied from the database 104 and stored in the storage unit 120, the management particle size determination unit 111 performs the work belonging to each process according to a plurality of analysis particle sizes determined in advance. Grouping into analysis particle size groups, calculate the average work time of the work belonging to each analysis particle size group.

Here, the analysis particle size is a combination of only the process or at least one of the process, the work content, the model, and the worker. For example, when there are three types of each of the process, the work content, the model, and the worker, if the analysis particle size is only the process, the work belonging to a certain process is grouped into one analysis particle size group Become. Also, if the analysis particle size is a process and a model, operations belonging to a certain process will be grouped into three analysis particle size groups.

Further, the control particle size determination unit 111 calculates the degree of variation of the working time of the work belonging to each analysis particle size group, and determines an analysis particle size that minimizes the calculated degree of variation as the control particle size in the process. For example, the control particle size determination unit 111 calculates an absolute value error between the operation time of the work belonging to each analysis particle size group and the average operation time, calculates a sum of them, and an analysis in which the sum of the absolute value errors is minimized. The particle size is determined to the control particle size in the process. Furthermore, the management granularity determination unit 111 stores the management granularity in the management granularity information 122 (details will be described later) stored in the storage unit 120.

In the present embodiment, the items of the control particle size and analysis particle size for each process are four items of process, work content, model, and worker, but other items such as materials may be added.

The evaluation value calculation unit 112 detects movement line information representing the movement route of the worker in each work from the moving image file obtained by imaging the work, which is included in the work record information 121 (details described later) stored in the storage unit 120. Then, the detection result is stored in flow line information 123 (details will be described later) stored in the storage unit 120. Further, the evaluation value calculation unit 112 calculates, based on the work record information 121 and the flow line information 123, a work time, a flow line distance, and a non-operation ratio, which are evaluation indices of the work. Furthermore, the evaluation value calculation unit 112 stores the calculated evaluation values in the evaluation value information 124 (details will be described later) stored in the storage unit 120.

The work classification unit 113 groups the work into a management particle size group according to the management particle size of each process based on the work performance information 121 stored in the storage unit 120, the management particle size information 122, and the evaluation value information 124. Select excellent work from among the above. Also, the work classification unit 113 stores the selected excellent work in the work classification information 125 (details will be described later) stored in the storage unit 120.

The work improvement point extraction unit 114 divides the angle of view of a moving image file obtained by imaging a work, which is included in the work result information 121 stored in the storage unit 120, into a plurality of work areas. Further, the work improvement point extraction unit 114 becomes an evaluation index of each work for each divided work area based on the work performance information 121, the management particle size information 122, and the flow line information 123 stored in the storage unit 120. Calculate the evaluation value. Furthermore, the work improvement point extraction unit 114 stores the evaluation value for each work area in the area-by-area evaluation value information 126 stored in the storage unit 120. Further, the work improvement point extraction unit 114 extracts work improvement points based on the work classification information 125 stored in the storage unit 120, the area-by-area evaluation value information 126, and the work improvement point library information 128. Furthermore, the work improvement point extraction unit 114 stores the extracted improvement points in the work improvement point information 127 stored in the storage unit 120.

The storage unit 120 is information necessary for work analysis, specifically, work record information 121, management particle size information 122, flow line information 123, evaluation value information 124, work classification information 125, area-by-area evaluation value information 126, work The improvement point information 127 and the work improvement point library information 128 are stored.

The input unit 130 receives operation information transmitted from the user terminal 103 via the network 102 and notifies the operation unit 110 of the operation information. Further, the input unit 130 receives the work record information supplied from the database 104 via the network 102, and adds it to the work record information 121 stored in the storage unit 120. Furthermore, according to the operation information of editing to the work improvement point library information 128 among the operation information transmitted from the user terminal 103, the input unit 130 stores the work improvement point library information 128 stored in the storage unit 120. Change

The output unit 140 (corresponding to the presentation control unit of the present invention) causes the display of the user terminal 103 to display an output screen 1300 (FIG. 14) representing the analysis result of the work. Further, the output unit 140 causes the display of the user terminal 103 to display the editing screen 1500 (FIG. 16) of the work improvement point library information 128.

Next, FIG. 2 shows an example of the data structure of the work record information 121. The work record information 121 stores information on a plurality of already performed works.

The work record information 121 is composed of a plurality of records corresponding to each work, and each record is composed of a work ID field 1211, a process field 1212, a work content field 1213, a model field 1214, and a worker field 1215. , A start time field 1216, an end time field 1217, and a moving image file field 1218.

The work ID field 1211 stores work ID (Identification) information for identifying each work. Process information is stored in the process field 1212. Here, the process information is information indicating which of a plurality of processes sequentially performed in a series of operations for manufacturing a product.

The work content field 1213 stores information representing the work content of the process (hereinafter, referred to as the process) represented by the process information stored in the process field 1212. A plurality of different work contents may exist for the same process.

The model field 1214 stores information indicating the model of the product manufactured in the process. The worker field 1215 stores information representing the worker who is in charge of the process. The start time field 1216 stores the work start time. The end time field 1217 stores the end time of the work. The moving image file field 1218 stores a moving image file obtained by capturing an operation. The frame rate of the moving image file may be, for example, about 1 fps (frames per second), but may be a higher frame rate such as 30 fps.

In the example of FIG. 2, for example, “work 1” in the process field 1212, “work 1” in the work content field 1213, and “machine 1” in the machine field 1214 for the record of work ID = work 1, The “worker 1” is stored in the worker field 1215. Also, "2017/4/2 9:00" is stored in the start time field 1216, "2017/4/2 9:30" is stored in the end time field 1217, and "moving image file ./movie 1" is stored in the moving image file field 1218. It is done.

FIG. 3 shows an example of the data structure of the control particle size information 122. As shown in FIG. The control particle size information 122 stores information indicating the control particle size of each process.

The control granularity information 122 is composed of a plurality of records, and each record has a process field 1221, a work content field 1222 for representing the control granularity, a model field 1223, and a worker field 1224.

The process field 1221 stores process information representing a process.

In the work content field 1222, information as to whether the work content is adopted or not is stored as the control granularity of the process. Specifically, “○” is stored when the work content is adopted as the management granularity, and “−” is stored when it is not adopted.

The model field 1223 stores information as to whether the model is adopted as the control granularity of the process. Specifically, “○” is stored when a model is adopted as the management granularity, and “−” is stored when not adopted.

In the worker field 1224, information as to whether the worker is adopted or not is stored as the control granularity of the process. Specifically, “○” is stored when a worker is adopted as the control granularity, and “−” is stored when it is not adopted.

In the example of FIG. 3, for example, as the control granularity of the process 1, none of the work content, the model, and the worker is employed. In this case, the control particle size of step 1 represents that a step is adopted. Thus, all operations belonging to step 1 are grouped into the same control granularity group and analyzed.

Also, for example, the control granularity of step 2 indicates that a model is adopted. Therefore, the operations belonging to the process 2 are grouped and analyzed in different control granularity groups for each model.

FIG. 4 shows an example of the data structure of the flow line information 123. The flow line information 123 stores information on the flow line of the worker in each work.

The flow line information 123 is composed of a plurality of records, and each record is composed of a work ID field 1231, a frame field 1232, an X coordinate field 1233, and a Y coordinate field 1234.

The work ID field 1231 stores a work ID for identifying each work. The frame field 1232 stores the frame numbers of the frames making up the moving image file. The X coordinate field 1233 and the Y coordinate field 1234 store the X coordinate and the Y coordinate of the barycentric position of the worker in the frame.

The example of FIG. 4 represents the X and Y coordinates of the center of gravity of the worker in each frame of the moving image file of the work 1. For example, the X and Y coordinates of the frame 1 are (29, 16). The coordinates represent (25, 10).

FIG. 5 shows an example of the data structure of the evaluation value information 124. As shown in FIG. The evaluation value information 124 stores evaluation values of each of a plurality of evaluation indexes of each work.

The evaluation value information 124 is composed of a plurality of records, and each record has a work ID field 1241, a work time field 1242, a flow line distance field 1243, and a non-operation ratio field 1244.

The work ID field 1241 stores a work ID for identifying each work. The working time field 1242 stores working time as the evaluation value of the evaluation index. The flow line distance field 1243 stores the flow line distance of the worker as the evaluation value of the evaluation index. The non-operation ratio field 1244 stores the non-operation ratio as the evaluation value of the evaluation index.

In the example of FIG. 5, for example, the evaluation index of work 1 is 30 minutes, the working distance is 5 m, the non-operation ratio is 10%, and the evaluation index of work 2 is 50 minutes of work time It represents that the line distance is 7 m and the non-operation ratio is 15%.

FIG. 6 shows an example of the data structure of the work classification information 125. The work classification information 125 stores information on excellent work at the control granularity of each process.

The work classification information 125 is composed of a plurality of records, and each record has a process field 1251, a work content field 1252, a model field 1253, a worker field 1254, and a work ID field 1255.

The process field 1251 stores process information representing a process. The work content field 1252 stores information on the work content among the control granularity of the process. If no work content is adopted for the control granularity of the process, “-” is stored in the work content field 1252. The model field 1253 stores information on the model of the control granularity of the process. If a model is not adopted as the control granularity of the process, “-” is stored in the model field 1253. The worker field 1254 stores information on the worker in the control granularity of the process. Note that “-” is stored in the worker field 1254 when the worker is not employed for the control granularity of the process.

The work ID field 1255 stores a work ID representing an excellent work in the control granularity of the process.

In the example of FIG. 6, for example, the control granularity of the process 1 is a process, and the excellent work of the work grouped into the control grain size group of the process 1 is a work 1. Further, for example, the control granularity of the process 2 is a model, and the excellent work of the work grouped into the control granularity group of the process 2 and the model 1 represents the work 3 and the work 5. Furthermore, the excellent work of the work grouped into the control granularity group of the process 2 and the model 2 represents the work 11.

FIG. 7 shows an example of the data structure of the evaluation value information 126 classified by area. The area-by-area evaluation value information 126 stores evaluation value information which is an evaluation index of each work collected for each work area.

The area-by-area evaluation value information 126 is composed of a plurality of records, and each record includes a work ID field 1261, a work area field 1262, an extraction start time field 1263, an extraction end time field 1264, and a work time field. 12 has a flow distance field 1266 and a non-operating ratio field 1267.

The work ID field 1261 stores a work ID for identifying each work. The work area field 1262 stores information representing a work area. The extraction start time field 1263 stores the start time of the work in the work area. An extraction end time field 1264 stores the end time of the work in the work area. The working time field 1265 stores working time in the working area. The movement distance of the worker in the work area is stored in the movement distance field 1266. The non-operation ratio field 1267 stores the non-operation ratio (described in detail later).

In the example of FIG. 7, the working time in the working area 2 of the working 1 is 5 minutes from 9:10 to 9:15 on 2017/4/2, the flow line distance is 2 m, and the non-operating ratio is 10%. Represents

FIG. 8 shows an example of the data structure of the work improvement point information 127. The work improvement point information 127 stores information on the work improvement point extracted at the control granularity of each process.

The work improvement point information 127 includes a plurality of records, and each record includes a work ID field 1271, a work area field 1272, an improvement point field 1273, an extraction start time field 1274, and an extraction end time field 1275. And.

The work ID field 1271 stores the work ID of the work whose improvement point has been extracted. The work area field 1272 stores information indicating the work area targeted for the improvement point. The improvement point field 1273 stores specific contents of the improvement point. The extraction start time field 1274 stores the extraction start time of the improvement point. The extraction end time field 1275 stores the extraction end time of the improvement point.

In the example of FIG. 8, in the work area 4 of the work 2, it is indicated that “the hand is flattering” is extracted between 8:00 and 8:15 of 2017/4/1 as the improvement point. Further, in the work area 2 of the work 7, it is indicated that “moving distance excess” is extracted between 13:00 and 13:10 of 2017/5/1 as an improvement point.

FIG. 9 shows an example of the data structure of the work improvement point library information 128. In the work improvement point library information 128, information to be referred to when extracting improvement points from each process is stored in advance. However, the work improvement point library information 128 can be newly registered or corrected by the user.

The work improvement point library information 128 is composed of a plurality of records, and each record includes a process field 1281, a work content field 1282, a model field 1283, a worker field 1284, an improvement point field 1285, and a work It has a time field 1286, a flow distance field 1287, and a non-operating rate field 1288.

The process field 1281 stores process information representing a process. The work content field 1282 stores information on the work content among the control granularity of the process. If no work content is adopted as the control granularity of the process, “-” is stored in the work content field 1282. The model field 1283 stores information on the model of the control granularity of the process. If a model is not adopted as the control granularity of the process, “-” is stored in the model field 1283. The worker field 1284 stores information on the worker in the control granularity of the process. When the worker is not employed for the control granularity of the process, “-” is stored in the worker field 1284.

The improvement point field 1285 stores the contents of the improvement point to be extracted. The working time field 1286 stores a threshold value of the difference between working time of excellent work and non-good work, which is referred to when extracting the improvement point. In the movement distance field 1287, a threshold of the difference between the movement distance between the excellent operation and the non- excellent operation, which is referred to when extracting the improvement point, is stored. In the non-operation ratio field 1288, a threshold of the difference between non-operation ratio of excellent work and non- excellent operation, which is referred to when extracting the improvement point, is stored.

In the example of FIG. 9, the condition that “Tatsuke at work” is extracted as an improvement point from the process 1 indicates that the difference in working time with the excellent work is 10 minutes or more. Further, the condition that “moving distance excess” is extracted as the improvement point from the process 1 represents that the difference in moving distance from the excellent work is 3 m or more.

<Work analysis processing by work analysis system 100>
Next, FIG. 10 is a flowchart illustrating an example of the task analysis process by the task analysis system 100.

This work analysis process is premised on the fact that a predetermined number of work record information is recorded in the database 104, and is started, for example, in response to a start command from the user.

First, the user terminal 103 receives an operation input from a user specifying a process to be analyzed, and transmits the operation information to the work analysis apparatus 101 via the network 102 (step S11). Next, the input unit 130 of the work analysis apparatus 101 having received the operation information acquires all the work record information corresponding to the process represented by the operation information from the database 104, and the work stored in the storage unit 120. It stores in the track record information 121 (step S12).

Next, the management particle size determination unit 111 of the calculation unit 110 determines the management particle size for the process represented by the operation information transmitted from the user terminal 103 based on the work record information 121 of the storage unit 120, and the storage unit 120 stores It stores in the management particle size information 122 which has been done (step S13).

Details of the process of step S13 will be described. The management particle size determination unit 111 reads a record matching the analysis particle size determined in advance from the work record information 121 stored in the storage unit 120, and calculates the work time of each record from the start time and the end time. For example, when the process to be analyzed is process 1 and the analysis particle size is a process, the record in which process 1 is stored in process field 1212 of work record information 121 is read, and start time field 1216 and end time field of each record The difference between the times stored in 1217 and 1217 is calculated as the working time. Further, for example, when the analysis target process is process 2 and the analysis particle size is the process and model, the record in which process 2 is stored in process field 1212 of work record information 121 is read, and is further stored in model field 1214 The analysis particle size group is grouped for each model ID, and the difference between the times stored in the start time field 1216 and the end time field 1217 of each record is calculated as the operation time for each analysis particle size group.

FIG. 11 visualizes the degree of variance of the calculated working time of each record, and in the process 1 and the process 2, the analysis particle size 1 grouped by only the process and the analysis particle size 2 grouped by combining the process and the model Is a scatter plot in which the working time of the work grouped into each analysis particle size group is plotted. The horizontal axis in the figure represents the work day, and the vertical axis represents the work time. Further, thick frame lines in FIG. 11 indicate the appropriate ones as the control particle sizes of the analysis particle size 1 and the analysis particle size 2 in each of the process 1 and the process 2 (the reason will be described later).

Next, the control particle size determination unit 111 quantifies the degree of variation of the operation time of the work grouped into the analysis particle size group. Specifically, the management particle size determination unit 111 calculates an average operation time for each analysis particle size group.

However, there is a possibility that incomplete data is included in the work record information 121 due to an input error of the start time or the end time at the time of registration. Therefore, the control particle size determination unit 111 removes outliers that may be defective data from the work record information 121.

Although there are several methods for removing outliers, in the present embodiment, histogram creation and Smirnov-Grabbs test are utilized. First, create a histogram of work time for each work. When creating a histogram, the number of histogram bins is determined according to the Stargest equation, but the number of histogram bins may be determined by other methods. After that, it is determined by the Smirnov-Grabs test whether or not the generated histogram contains an outlier. If no outlier is contained, the average value of the histogram is used as the average operation time. Conversely, if outliers are included, create the histogram again. Specifically, the records in the range where the frequency of the histogram is the highest and the ranges before and after it are extracted, and a histogram is created again using the extracted data, and the Smirnov-Grabs test is performed. Thereafter, the same processing is repeated until the generated histogram contains no outliers. When creating the histogram again, instead of extracting records in the range where the histogram frequency is the highest and the range before and after it, other ranges may be extracted, such as extracting only the range where the frequency is the highest. You may

In addition, you may utilize methods (The method based on a box and whiskers chart etc.) other than the method mentioned above for removal of an outlier.

FIG. 12 shows an example of the calculation result of the average operation time at each analysis particle size after removing the outliers. Similar to FIG. 11, FIG. 12 shows calculation results of the average operation time for the analysis particle size 1 of only the process and the analysis particle size 2 in which the process and the model are combined in the process 1 and the process 2. The average working time at each analysis particle size shown in FIG. 12 is plotted as a dotted line on the scatter plot of FIG. The average work time in analysis particle size 1 of step 1 is 10, the average work time of model 1 in analysis particle size 2 of step 1 is 11, the average work time of model 2 is 9 and the average work time of model 3 is 12 . In addition, the average working time in analysis particle size 1 of step 2 is 20, the average working time of model 1 in analysis particle size 2 of step 2 is 25, the average working time of model 2 is 35, and the average working time of model 3 is 21 It is.

Next, in each analysis particle size, the control particle size determination unit 111 determines, as the control particle size, the analysis particle size at which the variation of the operation time is minimized. Specifically, the control particle size determination unit 111 determines, as the control particle size of the process, an analysis particle size that minimizes the total sum of absolute value errors between the calculated average operation time and the operation time of each record. In addition, the calculation method of the dispersion degree of working time may be limited to the specific example mentioned above. For example, variance, standard deviation, etc. may be calculated.

If the sum of the absolute value errors is equal between the analysis particle sizes (in this case, between the analysis particle size 1 and the analysis particle size 2), the control particle size is determined as the one with the larger analysis particle size. The method of adoption will be described in detail with reference to FIG.

FIG. 13 shows the sum of absolute value errors of analysis particle size 1 and analysis particle size 2 in step 1 and step 2. In the case of step 1, the sum of absolute value errors of analysis particle size 1 is 300, and the sum of absolute value errors of analysis particle size 2 is also 300 (= 120 + 100 + 80). Therefore, in the case of step 1, since the sum of absolute value errors is equal between the analysis particle size 1 and the analysis particle size 2, the analysis particle size 1 of the coarser analysis particle size is determined as the control particle size. In the case of step 2, the sum of absolute value errors of analysis particle size 1 is 400, and the sum of absolute value errors of analysis particle size 2 is 350 (= 200 + 50 + 100). Therefore, in the case of the step 2, the analysis particle size 2 is determined as the control particle size because the total of the absolute value errors is smaller in the analysis particle size 2 than in the analysis particle size 1.

Finally, the management granularity determination unit 111 stores the determined management granularity in the management granularity information 122 (FIG. 3) stored in the storage unit 120. For example, in the case of FIG. 11, “process 1” is stored in the process field 1221 of the control particle size information 122 as a record corresponding to the process 1, and “-” is stored in the work content field 1222, the machine type field 1223 and the worker field 1224. Store Similarly, "process 2" is stored in the process field 1221 of the control granularity information 122 as a record corresponding to the process 2, "-" is stored in the work content field 1222 and the worker field 1224, and "type" is stored in the machine type field 1223. Stores "."

This is the end of the detailed description of the process of step S13. It returns to the description of the work analysis process of FIG.

Next, based on the work record information 121 stored in the storage unit 120, the evaluation value calculation unit 112 calculates and calculates the working time, the flow line distance, and the non-operation ratio, which are the evaluation values of the evaluation index of each work. The obtained evaluation value is stored in the evaluation value information 124 stored in the storage unit 120 (step S14).

In the present embodiment, three items of work time, flow distance and non-operation ratio are adopted as work evaluation index, but at least two of work time, flow distance and non-operation ratio May be adopted. Furthermore, in addition to the above three items, for example, time of each posture (standing, squatting, etc.), smoothness of movement, time of talking, movement of eyes, etc. may be adopted as an evaluation index.

Details of the process of step S14 will be described. First, in order to calculate an evaluation value, the evaluation value calculation unit 112 creates a flow line data by reading a moving image file of each operation from the operation result information 121 stored in the storage unit 120 and performing image analysis. Do. Specifically, the evaluation value calculation unit 112 searches the worker on each frame of the moving image file read from the work record information 121, and acquires the coordinates of the worker's center of gravity. As a method of searching for the worker, for example, a method is employed in which the feature of the worker is learned by machine learning in advance and the learning result and the image of each frame are compared, but other methods may be used.

Thereafter, the evaluation value calculation unit 112 stores the created flow line data in the flow line information 123 stored in the storage unit 120. Next, the evaluation value calculation unit 112 calculates the working time, the flow line distance, and the non-operation ratio based on the work record information 121 and the flow line information 123.

The work time is calculated by calculating the difference between the start time and the finish time of each work in the work record information 121. With regard to the flow line distance, the flow line distance of each operation is calculated by adding up the amount of change of the barycentric coordinates of the workers between the frames of the flow line information 123. With regard to the non-operation ratio, the flow line information 123 of each work detects the time spent in the previously designated work area (area where work is not performed), and the ratio of the detected time to the work time is the non-operation ratio Calculated as Finally, the evaluation value calculation unit 112 stores the calculated work time, flow line distance and non-operation ratio in the evaluation value information 124 stored in the storage unit 120.

This is the end of the detailed description of the process of step S14. It returns to the description of the work analysis process of FIG. Next, the work classification unit 113 performs work classification to classify excellent work from the work matching the process designated in step S11 (step S15).

The details of the process of step S15 will be described. First, the work classification unit 113 acquires the management grain size of the process with reference to the management grain size information 122 (FIG. 3), and acquires from the work record information 121 a record that matches the acquired management grain size. Further, the work classification unit 113 refers to the work ID field 1211 of the acquired record to acquire the work ID matching the management granularity of the process, and groups the work ID into the management granularity group. Furthermore, the work classification unit 113 acquires, from the evaluation value information 124 (FIG. 5), a record that matches the work ID belonging to each management granularity group. Furthermore, the work classification unit 113 refers to the records acquired from the evaluation value information 124, and selects candidate records that can be excellent work in each evaluation value of work time, flow distance and non-operation ratio.

For example, when the process designated in step S11 is process 1, it is acquired that the particle size is only the process from the control particle size information 122, and then the work ID matching the process 1 from the work record information 121 (FIG. In the case, operations 1 to 7) are acquired and grouped into management granularity groups. Further, from the evaluation value information 124, records matching the tasks 1 to 7 are obtained.

Further, for example, when the process designated in step S11 is process 2, the particle size is acquired from the management particle size information 122 as the process and the model, and next, the operation record information 121 matches the process 2 and the model 2. A work ID (in the case of FIG. 2, work 11) is acquired. Further, from the evaluation value information 124, a record matching the operation 11 (not shown in FIG. 5) is acquired.

Hereinafter, a method of selecting records that can be candidates for excellent work based on the evaluation value of work time will be described.

First, the work classification unit 113 calculates the average work time based on the record acquired from the evaluation value information 124. In addition, as a method of calculating the average work time here, the average work time is calculated after removing outliers from the work time, similarly to the processing in the management particle size determination unit 111. Next, the work classification unit 113 selects a record whose work time is equal to or less than the average work time as a candidate for excellent work.

In addition, the work classification unit 113 similarly selects records having a flow line distance equal to or less than the average flow line distance for evaluation values other than the work time as candidates for excellent work, and the non-operation ratio averages non-operation Select a record that is less than the percentage as a candidate for excellent work.

Next, the work classification unit 113 selects a record selected as a candidate for excellent work in all evaluation indexes (work time, flow distance and non-operation ratio) as excellent work. If a plurality of records are candidates for excellent work in all evaluation values, the corresponding multiple records are selected as excellent work.

Finally, the work classification unit 113 stores the selected excellent work record in the work classification information 125 stored in the storage unit 120. In the case of process 1 of FIG. 3, the control particle size is only the process, so the work classification unit 113 stores “process 1” in the process field 1251, and the work content field 1252, the model field 1253, and the worker field 1254. And “work 1” selected as an excellent work in the work ID field 1255. Further, in the case of step 2 of FIG. 3, since the control granularity is the step and model, the work classification unit 113 stores “step 2” in the step field 1251 and stores “model 1” of the model field 1253 , “-” Is stored in the work content field 1252 and the worker field 1254, and “work 3” selected as the excellent work is stored in the work ID field 1255. Further, the work classification unit 113 stores “step 2” in the step field 1251, stores “model 1” in the model field 1253, and stores “−” in the work content field 1252 and the worker field 1254. The "work 5" selected as the excellent work is stored in the work ID field 1255. Further, the work classification unit 113 stores “step 2” in the step field 1251, stores “model 2” in the model field 1253, and stores “−” in the work content field 1252 and the worker field 1254. The "work 11" selected as the excellent work is stored in the work ID field 1255.

This is the end of the detailed description of the process of step S15. It returns to the description of the work analysis process of FIG.

Next, the work improvement point extraction unit 114 performs steps based on the work record information 121, the management particle size information 122, the flow line information 123, the work classification information 125, and the work improvement point library information 128 stored in the storage unit 120. The improvement point of the work in the management granularity of the process designated in S11 is extracted, and the extracted improvement point is stored in the work improvement point information 127 stored in the storage unit 120 (step S16).

Details of the process of step S16 will be described. First, the work improvement point extraction unit 114 refers to the management particle size information 122 to acquire the management particle size of the process specified in step S11, and acquires a record matching the acquired management particle size from the work record information 121 Identify the ID.

For example, in the case of process 1 of FIG. 3, the management granularity is only the process, and the work improvement point extraction unit 114 acquires a record matching the process 1 from the work result information 121 and acquires the work ID (in the case of FIG. 3). Task ID1 to task ID7) are identified.

Next, the work improvement point extraction unit 114 acquires a record that matches the specified work ID from the flow line information 123, and based on the X, Y coordinates of the center of gravity of the worker in the acquired record, the moving image file of the process The angle of view of is divided into multiple work areas.

As a method of dividing the angle of view of the moving image file into the work area, the Starge's equation for determining the number of bins of the histogram is utilized, and from the data of the X coordinate stored in the flow line information 123, the horizontal direction of the work area The number of divisions of the work area in the vertical direction is determined from the Y coordinate data stored in the flow line information. Also, the work area may be divided according to the input from the user using the user terminal 103.

Next, the work improvement point extraction unit 114 calculates, for each work area, the work time, the flow distance, and the non-operation ratio, which are evaluation values of the work, in the management granularity of the process. First, the work improvement point extraction unit 114 acquires, from the management particle size information 122 and the flow line information 123, a record that matches the management particle size of the process. For example, in the case of step 1 of FIG. 3, since the control granularity is only the step, the work improvement point extraction unit 114 acquires a record in which the step information matches the step 1 from the work record information 121 and the flow line information 123. Do.

As for the working time by working area, the range of the working area and the barycentric coordinates of the worker stored in the flow line information 123 are compared, and the worker specifies the frame included in the working area. The work time in the work area is calculated from the specified frame and the number of frame rates. Also, the start time and the end time of the work in each work area are acquired from the specified frame.

Regarding the flow line distance for each work area, the range of the work area and the coordinates of the center of gravity of the worker stored in the flow line information 123 are compared, and the worker specifies the frame included in the work area Do. After that, the flow line distance in the work area is calculated from the change distance of the flow line between the specified frames.

With regard to the non-operation ratio for each work area, the range of the work area and the coordinates of the barycenter of the worker stored in the flow line information 123 are compared, and the time spent in the non-work area in the work area is compared. Calculate the ratio of work area to work time and the non-operation ratio of each work area.

Next, the work improvement point extraction unit 114 stores the calculated work time, flow line distance, and non-operation ratio for each work area, the start time of the work in the work area, and the end time in the evaluation value information 126 for each area. Do.

Next, the work improvement point extraction unit 114, based on the work record information 121, the management grain size information 122, the work classification information 125, the evaluation value information by area 126, and the work improvement point library information 128, in the management granularity of the process. Extract improvement points of non-excellent work that were not classified as excellent work. Specifically, first, the work improvement point extraction unit 114 acquires excellent work in the control granularity of the process based on the work classification information 125. For example, in the case of process 1 of FIG. 3, the work improvement point extraction unit 114 acquires the work 1 as the excellent work in the process 1 from the work classification information 125.

Next, the work improvement point extraction unit 114 calculates the difference between the evaluation values of the excellent work and the non- excellent work based on the work record information 121 and the evaluation value information by area 126, and the difference between the evaluation values improves the work. It is determined whether it is equal to or more than the threshold of the evaluation value registered in the point library information 128, and an improvement point is extracted.

For example, in the case of process 1 of FIG. 3, the work improvement point extraction unit 114 calculates each evaluation value of work 1 which is excellent work and work 7 which is non- excellent work from the evaluation value information 126 classified by area (FIG. 7). Calculate the difference. In this case, the difference between working time of work 1 which is excellent work and work 7 which is non-good work is calculated as “5 minutes”, the difference of flow line distance is “4 m” and the difference of non-operating ratio is “10%” Be done. Furthermore, the threshold of each evaluation value is acquired from the work improvement point library information 128 (FIG. 8), and it is determined that the difference "4 m" in flow line distance is equal to or more than the threshold "3 m". , “Travel distance exceeded” is selected.

Next, the work improvement point extraction unit 114 stores the extracted improvement points in the work improvement point information 127 stored in the storage unit 120. For example, in the case of work 7 described above, the work improvement point extraction unit 114 stores “work 7” in the work ID field 1271 of the work improvement point information 127, stores “area 2” in the work area field 1272, and improves “Moved distance exceeded” is stored in the point field 1273. Furthermore, the work improvement point extraction unit 114 stores “2017/5/1 13:00” in the extraction start time field 1274 of the work improvement point information 127 and “2017/5/13 13: in the extraction end time field 1275. Store 10 ".

This is the end of the detailed description of the process of step S16. It returns to the description of the work analysis process of FIG.

Finally, the output unit 140 generates an output screen 1300 (FIG. 14) representing the result of the work analysis based on each information stored in the storage unit 120, and outputs the generated output screen 1300 to the user terminal 103 via the network 102. Further, the output unit 140 updates the output screen 1300 as needed in response to an operation from the user on the output screen 1300 and outputs the updated screen 1300 to the user terminal 103. The user terminal 103 presents it to the user by displaying the output screen 1300 on the display (step S17). Thus, the task analysis process by the task analysis system 100 is completed.

Next, FIG. 14 illustrates a display example of the output screen 1300 displayed on the user terminal 103. The output screen 1300 includes a process information selection field 1301, a control particle size display field 1302, an analysis object selection field 1303, an excellent work display field 1304, a work improvement point display field 1305, and a work improvement point library information display field 1306. , Library correction button 1307. Further, the output screen 1300 has an excellent work moving image display field 1308 and an extracted work moving image display field 1309.

In the process information selection field 1301, the user can select a process for displaying the analysis result. The control particle size display column 1302 displays the control particle size of the process (the process selected in the process information selection column 1301). In the analysis object selection column 1303, the user can select an analysis object of management granularity for displaying the analysis result. When the control particle size of the process is a process, “-” is displayed in the control particle size display field 1302, and an analysis object of the control particle size can not be selected in the analysis object selection field 1303. Further, for example, when the management particle size of the process is a worker, “worker” is displayed in the management particle size display field 1302, and the worker can be selected in the analysis object selection field 1303.

The excellent work display column 1304 displays a work ID and a worker as a record of the excellent work in the process. In the work improvement point display column 1305, the work ID, the operator, the work area, the improvement point, the extraction start time, and the extraction end time Is displayed. In the work improvement point display column 1305, the user can select the displayed record, and the record surrounded by a bold line frame (in this case, work 7) is selected.

In the work improvement point library information display column 1306, the threshold of the evaluation index as the basis for the improvement point being extracted and the excellent work and the improvement point of the work selected by the user in the work improvement point display column 1305 are extracted The difference between the evaluated work and the evaluated work is displayed. The library correction button 1307 is a button for starting an editing process for newly registering a record in the work improvement point library information 128 or correcting a registered record, and editing when the library correction button 1307 is pressed. Screen 1500 (FIG. 15) is displayed.

In the excellent work moving picture display field 1308, a moving picture file of the excellent work in the process is reproduced and displayed. In the extracted work moving image display field 1309, a moving image file of the work selected by the user in the work improvement point display field 1305 is reproduced and displayed.

Next, FIG. 15 shows a display example of the editing screen 1500. The edit screen 1500 has a process information selection field 1501, a new registration reception unit 1502, a registration button 1503, a correction reception unit 1504, and a correction button 1505.

In the process information selection field 1501, the user can select the process of the record to be newly registered in the work improvement point library information 128 or the record to be corrected.

The user can input a record to be newly registered in the work improvement point library information 128 to the new registration reception unit 1502. When a new registration acceptance unit 1502 is used to newly register a record, there is a case where the control granularity of the process selected in the process information selection field 1501 is not set in advance. In that case, since the user needs to set the work content, the model, and the worker in combination, the user may set the control granularity with respect to the process based on the experience of the expert. If the management granularity set here is inappropriate, the management granularity may be corrected using a correction accepting unit 1504 described later, based on the result of the task analysis by the task analysis device 101.

The registration button 1503 can instruct the new registration reception unit 1502 to register the record input in the work improvement point library information 128.

The correction receiving unit 1504 can display the existing record in the work improvement point library information 128 so that the user can correct it. The correction button 1505 can reflect the correction input by the correction receiving unit 1504 in the work improvement point library information 128 when the user presses it.

Next, FIG. 16 is a flowchart illustrating an example of an editing process that can newly register or correct the work improvement point library information 128. This editing process is started in response to pressing of the library correction button 1307 on the output screen 1300, and the editing screen 1500 is displayed on the user terminal 103.

First, after the user selects a process in the process information selection field 1501 of the edit screen 1500, an input is made to the new registration reception unit 1502 and the registration button 1503 is pressed, or a correction reception unit 1504 is displayed. Then, input is performed and the correction button 1505 is pressed. In response to this, the user terminal 103 transmits the operation information to the input unit 130 of the work analysis apparatus 101 via the network 102 (step S21).

The input unit 130 that receives the transmitted operation information newly registers a record in the work improvement point library information 128 stored in the storage unit 120 based on the received operation information, or corrects an existing record. The result is stored (step S22). Thus, the editing process is ended.

As described above, according to the task analysis system 100 according to the present embodiment, the task analysis device 101 includes the management particle size determination unit 111, so that it is possible to determine an appropriate management particle size. Further, since the work analysis device 101 includes the evaluation value calculation unit 112, evaluation values of a plurality of different evaluation indexes can be calculated without relying on human power. Further, since the work analysis device 101 includes the work classification unit 113 and the work improvement point extraction unit 114, the work improvement point can be extracted based on the excellent work and the non- excellent work. Further, since the work improvement point extraction unit 114 extracts the work improvement point with reference to the work improvement point library information 128, the user adjusts the standard of the work improvement point by editing the work improvement point library information 128. be able to.

The work analysis apparatus 101 according to the above-described embodiment can be configured by hardware or can be realized by software. When the work analysis apparatus 101 is realized by software, a program that configures the software is installed in a computer. Here, the computer includes, for example, a general-purpose personal computer that can execute various functions by installing a computer incorporated in dedicated hardware and various programs.

FIG. 17 is a block diagram showing an example of a hardware configuration of a computer that realizes the work analysis apparatus 101 by a program.

In the computer 2000, a central processing unit (CPU) 2001, a read only memory (ROM) 2002, and a random access memory (RAM) 2003 are mutually connected by a bus 2004.

Further, an input / output interface 2005 is connected to the bus 2004. An input unit 2006, an output unit 2007, a storage unit 2008, a communication unit 2009, and a drive 2010 are connected to the input / output interface 2005.

The input unit 2006 includes a keyboard, a mouse, a microphone and the like. The output unit 2007 includes a display, a speaker, and the like. The storage unit 2008 includes a hard disk, a non-volatile memory, and the like. The communication unit 2009 includes a network interface and the like. The drive 2010 drives removable media 2011 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory.

In the computer 2000 configured as described above, for example, the CPU 2001 loads a program stored in the storage unit 2008 into the RAM 2003 via the input / output interface 2005 and the bus 2004 and executes the task analysis. The arithmetic unit 110, the input unit 130, and the output unit 140 of the device 101 are realized.

Further, the storage unit 120 of the work analysis apparatus 101 is realized by the storage unit 2008, the RAM 2003, or the removable medium 2011.

The program executed by the computer 2000 (CPU 2001) can be provided by being recorded on, for example, the removable medium 2011 as a package medium or the like. Also, the program can be provided via a wired or wireless transmission medium such as a local area network, the Internet, or digital satellite broadcasting.

In the computer 2000, the program can be installed in the storage unit 2008 via the input / output interface 2005 by attaching the removable media 2011 to the drive 2010. Also, the program can be received by the communication unit 2009 via a wired or wireless transmission medium and installed in the storage unit 2008. In addition, the program can be installed in advance in the ROM 2002 or the storage unit 2008.

The program executed by computer 2000 may be a program that performs processing in chronological order according to the order described in this specification, or in parallel, or when necessary, such as when a call is made. The program may be a program to be processed in

The effects described in the present specification are merely examples and are not limited, and other effects may be present.

The present invention is not limited to the embodiments described above, but includes various modifications. For example, the above-described embodiments are described in detail in order to explain the present invention in an easy-to-understand manner, and the present invention is not necessarily limited to those having all the described components. Also, part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of one embodiment. Moreover, it is possible to add, delete, and replace other configurations for part of the configurations of the respective embodiments.

Further, each of the configurations, functions, processing units, processing means, etc. described above may be realized by hardware, for example, by designing part or all of them with an integrated circuit. Further, each configuration, function, etc. described above may be realized by software by the processor interpreting and executing a program that realizes each function. Information such as a program, a table, and a file for realizing each function can be placed in a memory, a hard disk, a storage device such as a solid state drive (SSD), or a recording medium such as an IC card, an SD card, or a DVD. Further, control lines and information lines indicate what is considered to be necessary for the description, and not all control lines and information lines in the product are necessarily shown. In practice, almost all configurations may be considered to be mutually connected.

The present invention can be provided not only in a work analysis apparatus and a work analysis method, but also in various modes such as a system including a plurality of apparatuses and a computer readable program.

100 ... work analysis system, 101 ... work analysis device, 102 ... network, 103 ... user terminal, 104 ... database, 110 ... calculation unit, 111 ... management particle size determination unit , 112: evaluation value calculation unit, 113: work classification unit, 114: work improvement point extraction unit, 120: storage unit, 121: work performance information, 122: management particle size information , 123: flow line information, 124: evaluation value information, 125: work classification information, 126: evaluation value information by area, 127: work improvement point information, 128: work improvement Point library information, 130: input unit, 140: output unit, 1211: work ID field, 1212: process field, 1213: work content field, 1214 · · · Machine type field, 1215 ... worker field, 1216 ... start time field, 1217 ... end time field, 1218 ... movie file field, 1221 ... process field, 1222 ... work content Field, 1223 ... model field, 1224 ... worker field, 1231 ... work ID field, 1232 ... frame field, 1233 ... X coordinate field, 1234 ... Y coordinate field, 1241 · · · · · Work ID field, 1242 · · · Work time field, 1243 · · · · flow line distance field, 1244 · · · non-operation ratio field, 1251 · · · process field, 1252 · · · work content field, 1253 · · ·・ Model field, 1254 ・ ・ ・ Product Operator field 1255 Operation ID field 1261 Operation ID field 1262 Operation area field 1263 Extraction start time field 1264 Extraction end time field 1265 Operation Time field, 1266 ... flow line distance field, 1267 ... non-operation ratio field, 1271 ... work ID field, 1272 ... work area field, 1273 ... improvement point field, 1274 ... extraction Start time field, 1275 extraction end time field, 1281 step field, 1282 work content field, 1283 type field, 1284 operator field, 1285 improvement point field , 1286 ... at the time of work Between field, 1287 ... flow line distance field, 1288 ... non-operation ratio field, 1300 ... output screen, 1301 ... process information selection field, 1302 ... management granularity display field, 1303 ... Analysis target selection column, 1304 ... excellent work display column, 1305 ... work improvement point display column, 1306 ... work improvement point library information display column, 1307 ... library correction button, 1308 ... excellent work Moving image display field 1309 Extraction operation moving image display field 1500 Editing screen 1501 Process information selection field 1502 New registration acceptance section 1503 Registration button 1504 Correction reception unit, 1505 ... correction button, 2000 ... computer, 2001 ... CPU, 2002 ... ROM, 2003 · · RAM · 2004 · · · Bus · 2005 · · · I / O interface, 2006 · · · input unit, 2007 · · · output unit, 2008 · · · storage unit, 2009 · · · communication unit, 2010 · · · drive , 2011 ... removable media

Claims (13)

1. A control particle size determination unit that determines a control particle size for each process based on work record information in which information on a plurality of executed works is accumulated;
   An evaluation value calculation unit that calculates an evaluation value of each of a plurality of evaluation indices for each work based on the work performance information;
   Operations are grouped into management granularity groups in accordance with the determined management granularity for each process, and based on the evaluation value of each of the plurality of evaluation indices calculated for the operations, excellent from the plurality of operations belonging to each management granularity group A work classification unit that classifies work;
   A work improvement point extraction unit that extracts a work improvement point of a non- excellent work belonging to each management granularity group based on the evaluation value of the excellent work of each management granularity group;
   A work analysis apparatus comprising:
2. The operation analysis apparatus according to claim 1, wherein
   A storage unit that stores work improvement point library information in which management granularity, work improvement points, and the threshold of each evaluation index are associated with one another for each process;
   Equipped with
   The work improvement point extraction unit belongs to each of the control particle size groups based on the evaluation value of each of the classified excellent work, the evaluation value of each of the non- excellent work, and the work improvement point library information. A work analysis apparatus characterized by extracting the work improvement points of non- excellent work.
3. The work analysis apparatus according to claim 2, wherein
   The work improvement point library information can be edited by a user.
4. The operation analysis apparatus according to claim 1, wherein
   The control particle size determination unit groups operations into analysis particle size groups according to a plurality of analysis particle sizes determined in advance, and the management is selected from among the analysis particle sizes based on the degree of variation in operation time of the work belonging to each analysis particle size group. A work analysis apparatus characterized by determining a particle size.
5. The work analysis apparatus according to claim 4, wherein
   The control particle size determination unit excludes outliers of the operation time of the work belonging to each analysis particle size group of the work belonging to each analysis particle size group, and the analysis particle size based on the dispersion condition of the operation time excluding the outliers. A work analysis apparatus characterized by determining the control granularity among them.
6. The work analysis apparatus according to claim 4, wherein
   The control particle size determination unit groups operations into analysis particle size groups according to a plurality of analysis particle sizes determined in advance, and based on an absolute value error between an average operation time of each operation belonging to each analysis particle size group and each operation time A work analysis apparatus characterized by determining the control particle size from analysis particle sizes.
7. The operation analysis apparatus according to claim 1, wherein
   The work analysis apparatus, wherein the evaluation value calculation unit calculates at least two of a working time, a flow line distance, and a non-operation ratio as the evaluation value of the evaluation index.
8. The operation analysis apparatus according to claim 1, wherein
   The work classification unit sets a work superior to the average in each evaluation index as an excellent work candidate, and classifies the work regarded as the excellent work candidate in all of the plurality of evaluation indices into the excellent work. Work analysis device.
9. The operation analysis apparatus according to claim 1, wherein
   After excluding the outliers in each evaluation index, the work classification unit sets an operation superior to the average in each evaluation index as an excellent work candidate, and the operation in which all the plurality of evaluation indices are considered as the excellent work candidate A work analysis apparatus characterized in that the work is classified as the excellent work.
10. The operation analysis apparatus according to claim 1, wherein
    The work improvement point extraction unit divides the moving image included in the work performance information into a plurality of areas, and extracts the work improvement point based on the evaluation value of the excellent work for each divided area. Work analysis device.
11. The operation analysis apparatus according to claim 10, wherein
    The work analysis device, wherein the work improvement point extraction unit divides the moving image into a plurality of areas based on a movement range of a worker in the moving image included in the work result information.
12. The operation analysis apparatus according to claim 1, wherein
    A presentation control unit that causes the user to present the extracted work improvement points;
    A work analysis apparatus comprising:
13. It is a work analysis method of a work analysis device, and
    A control granularity determination step of determining management granularity for each process based on the work record information in which information on a plurality of executed works is accumulated;
    An evaluation value calculation step of calculating an evaluation value of each of a plurality of evaluation indicators for each work based on the work record information;
    Operations are grouped into management granularity groups in accordance with the determined management granularity for each process, and based on the evaluation value of each of the plurality of evaluation indices calculated for the operations, excellent from the plurality of operations belonging to each management granularity group An operation classification step for classifying operations;
    A work improvement point extraction step of extracting a work improvement point of a non- excellent work belonging to each of the control granularity groups based on the evaluation value of the excellent work of the control granularity groups;
    An operation analysis method characterized in that it comprises:

Images