JP7280496B2 - Information processing device, information processing method, and information processing program - Google Patents

Description

The present application relates to an information processing device, an information processing method, and an information processing program .

In a production line, a discrepancy occurs between the production plan and the actual production. Therefore, techniques for improving production lines using indicators such as KPI have been disclosed so as to improve production results (see, for example, Patent Documents 1 to 5).

JP 2018-195308 A JP 2017-162044 A JP-A-2009-3794 JP 2005-157819 A JP 2018-156346 A

In order to greatly improve the KPI, it is desirable to calculate the degree of correlation between each production performance parameter and the KPI for the production performance data. However, if the number of actual production data for calculating the degree of correlation is large, it takes time to perform arithmetic processing for calculating the degree of correlation.

In one aspect, an object of the present invention is to provide an information processing device, an information processing method, and an information processing program capable of reducing the number of actual production data for calculating the degree of correlation with KPIs. .

In one aspect, the information processing device includes a calculation unit that calculates a representative value for each production performance parameter for the accumulated production performance data, and a master table that defines directions for improving the production process. a selection unit that selects, for each actual result parameter , only the actual production data that improves the production process from among the actual production data smaller and larger than the representative value calculated by the calculation unit; and a calculation unit that calculates a degree of correlation with the KPI related to the production performance for each of the production performance parameters using the production performance data selected by.

It is possible to reduce the number of production performance data for calculating the degree of association with KPI.

(a) is a schematic diagram of a production line, and (b) is a diagram illustrating a case in which products with different work times and setup change times are sent to a plurality of production lines. FIG. 4 is a diagram illustrating the flow of digital planning processing; FIG. 4 is a diagram illustrating the flow of digital planning processing; FIG. 10 is a diagram illustrating the rate of change of KPI with respect to changes in production performance parameters; 1A is a block diagram illustrating an overall configuration of an information processing apparatus according to a first embodiment; FIG. 1B is a block diagram illustrating a hardware configuration of the information processing apparatus; FIG. FIG. 3 is a diagram illustrating a flowchart representing an example of processing executed by an information processing device; (a) is a diagram illustrating a master table, and (b) and (c) are diagrams illustrating a range of selected production performance data. It is a figure which illustrates the actual production data selected for every adjustment parameter.

Before describing the embodiments, an overview of improvement activities using key performance indicators (KPIs) will be described.

As an example, in a production line of a factory, multiple tasks are required to assemble a product, and humans and robots are arranged to perform the multiple tasks. FIG. 1(a) is a schematic diagram of a production line. In a production line, a product 1 to be produced is placed on a belt conveyor 2 . A product 1 is an unfinished product and is a work target of each process worker. A product 1 is sequentially conveyed to each process by a belt conveyor 2 . Human and robot workers are assigned to each process.

It is preferable that each work is performed within a set working time. However, the work may be prolonged, or the work may have to be interrupted (temporary stop) due to a malfunction of the production equipment, and the work may be performed beyond the work time. The actual time required for each worker to perform the work is called cycle time (CT). If the CT is longer than the determined work time, the production completion time will be delayed. Among CTs, the cycle time when the operator is a robot is called MCT. A shorter average value of CT and MCT is preferable because a shorter average value means better production. In addition, when the variation in CT and MCT is small, each work time is leveled, so it is preferable that the variation is small.

For example, an indicator light 3 is arranged in each process. The indicator lamp 3 is an indicator lamp for notifying an alarm. When a problem occurs in the work, the work is temporarily interrupted (suspended) and the indicator lamp 3 of the same process lights up. When the indicator lamp 3 lights up, the support worker can grasp the process requiring support. After that, the support worker supports the process. The shorter the average recovery time for recovering from the temporary stop, the less the impact on the production process, so the shorter the better. In addition, since each recovery time is leveled when the variation in recovery time is small, it is preferable that the variation is small. A lower frequency of temporary stop is preferable because a lower frequency means that a good production process is being carried out. From the viewpoint of shortening the production completion time, it is preferable that the number of support workers (number of resources) is large. In addition, when a plurality of temporary stops overlap, a good production process can be performed by considering the priority of the support order by the support workers.

Next, when performing high-mix low-volume production on a single production line, the types of products change many times. Therefore, every time the type of product is changed, the settings of the production equipment are changed. Such a setting change is called a "stage change". A changeover is sometimes referred to as a "setup". The shorter the changeover time required for the changeover, the better the production process is carried out. A small variation in the changeover time is preferable because each changeover time is leveled out when the variation is small.

FIG. 1(b) is a diagram illustrating a case where products with different work times and changeover times are sent to a plurality of production lines. The work is stopped while the changeover work is being performed. The changeover time differs from product to product. When one person performs a plurality of changeovers, the production completion time can be shortened by devising the order of priority of the changeover work. From the viewpoint of shortening the production completion time, it is preferable that the number of support workers (number of resources) is large. Moreover, when a plurality of changeover operations overlap, the production completion time can be shortened by considering the order of priority of the changeover by the support worker.

In such a complicated production line as described above, it is important to obtain a quantitative index for judging whether or not the production process is being carried out satisfactorily. Table 1 is a diagram illustrating KPIs in a production line. There are various types of KPIs. For example, KPIs include operational KPIs and quality KPIs. For example, as exemplified in Table 1, operational KPIs include load efficiency, production end time, overtime hours, non-operating hours, work duplication rate, and the like. Load efficiency is the ratio of operating time (time during which work is being done) to total production time. A higher load efficiency is preferable. The production end time is the time when production is completed. The earlier the production end time, the better. The overtime hours are the time obtained by subtracting the actual working hours from the actual working hours. Less overtime is preferable. The non-operating time is the work waiting occurrence time. Less non-working time is preferable. The work duplication rate is the average time when the number of work duplications is tallied for each hour. It is preferable that the work duplication rate is as small as possible.

In improvement activities based on KPIs, for example, production managers rely on their know-how and intuition to determine improvement measures that will improve KPIs, and deploy them at production sites. However, if the production manager is absent or his intuition is off, it will take time to repeat trial and error.

Therefore, it is conceivable to determine improvement measures through digital planning. Digital planning is to virtually verify the KPI change when the values of each production performance parameter are assigned.

FIG. 2 is a diagram illustrating the flow of digital planning processing. As illustrated in FIG. 2, first, plan data and performance data are entered. The plan data includes an input plan, changeover, and the like. An input plan is a plan for completing production, and includes the type of product to be produced (variety), the work required for each product, the work time determined for each work, the number of products, each production The order of input to the line, the timing of input to each production line, and the like. The changeover of plan data is the content of the work required for the changeover. The performance data includes production performance data such as operation information, changeover information, and error information. The operation information is statistics such as the average work time actually required for each work and variations. The changeover information is information about the changeover that was actually performed. The error information is information regarding malfunctions of each work such as suspension.

Digital planning then takes place. In digital planning, statistical processing is first performed on each piece of actual production data. Next, a virtual performance is constructed by adjusting the statistical values obtained by the statistical processing as production performance parameters. In the example of FIG. 2, it is assumed that the average work time is 5 minutes and the work time variation σ is 2 minutes when a specific person works on a product of type A001 for facility number #1. Build virtual performance data for changing these values. As a result, the order of priority of temporary stop restoration work, the order of priority of changeovers, and the like are determined, and the respective works and changeovers are rearranged. Next, KPIs are calculated for the sorted virtual performance data. The calculated KPI is output and displayed on a display device or the like. By comparing the KPI calculated from the actual production data with the KPI calculated from the virtual actual data, it is possible to determine whether the virtual actual data has been improved. If the virtual performance data is improved, it will be possible to determine what to improve.

However, in digital planning, the number of production performance parameters increases, and there is a problem that it is difficult to specify which production performance parameter should be adjusted to improve the KPI. Therefore, attention is focused on the correlation between each production performance parameter and KPI. As illustrated in FIG. 3, each production performance parameter and KPI have a monotonically increasing or monotonically decreasing relationship. Therefore, it is conceivable to calculate the correlation coefficient between each production performance parameter and KPI, and obtain the sensitivity (correlation degree) of each production performance parameter from the slope of the approximation line in that case. The greater the slope of the approximation line, the greater the degree of correlation. Once the correlation of each production performance parameter is obtained, the KPI can be greatly improved by adjusting the production performance parameter with a large correlation.

Therefore, as illustrated in FIG. 4, the rate of change in KPI with respect to the change in each production performance parameter is calculated as the degree of correlation. The production performance parameters are sorted in descending order of correlation. In the example of FIG. 4, the step change average has the highest degree of correlation. Therefore, by adjusting this stage change average, it becomes possible to efficiently improve the KPI.

In order to calculate the degree of correlation with KPI, calculation is performed using a large amount of actual production data. If the amount of actual production data required to calculate the degree of correlation is large, it will take a long time for arithmetic processing.

In the following embodiment, information processing that can shorten the time required to calculate the correlation between each production performance parameter and the KPI by reducing the number of production performance data for calculating the correlation with the KPI. A device, a production information calculation method, and a production information calculation program will be described.

FIG. 5A is a block diagram illustrating the overall configuration of the information processing apparatus 100 according to the first embodiment. As illustrated in FIG. 5A, the information processing apparatus 100 includes a performance data accumulation unit 10, a statistical processing unit 20, a statistical data accumulation unit 30, a representative value calculation unit 40, a representative value data accumulation unit 50, a data selection unit 60, KPI change amount calculation unit 70, correlation degree accumulation unit 80, and the like.

FIG. 5B is a block diagram illustrating the hardware configuration of the information processing apparatus 100. As illustrated in FIG. As illustrated in FIG. 5B, the information processing apparatus 100 includes a CPU 101, a RAM 102, a storage device 103, a display device 104, and the like. A CPU (Central Processing Unit) 101 is a central processing unit.

CPU 101 includes one or more cores. A RAM (Random Access Memory) 102 is a volatile memory that temporarily stores programs executed by the CPU 101, data processed by the CPU 101, and the like. The storage device 103 is a non-volatile storage device. As the storage device 103, for example, a ROM (Read Only Memory), a solid state drive (SSD) such as a flash memory, a hard disk driven by a hard disk drive, or the like can be used. The storage device 103 stores an information processing program. The display device 104 is a liquid crystal display, an electroluminescence panel, or the like, and displays calculation results and the like. In this embodiment, each part of the information processing apparatus 100 is realized by executing a program, but hardware such as a dedicated circuit may be used.

FIG. 6 is a diagram illustrating a flowchart representing an example of processing executed by the information processing apparatus 100. As shown in FIG. As illustrated in FIG. 6, the statistical processing unit 20 extracts actual production data from the actual production data storage unit 10 that stores various actual production data of the production line (step S1).

Next, the statistical processing unit 20 categorizes the actual production data by status, production apparatus, worker, and product type (step S2). In this embodiment, the status is a production performance parameter, and includes MCT [min], changeover time [min], pause recovery time [min], pause occurrence frequency [times], number of resources [persons], Consideration of work priority, etc.

Next, the statistical processing unit 20 performs statistical processing for each category categorized in step S2 (step S3). For example, the statistical processing unit 20 calculates maximum value max, minimum value min, average value ave, standard deviation σ, and the like. The calculated statistical data are accumulated in the statistical data accumulation unit 30 .

Next, the representative value calculator 40 calculates a representative value for each actual production parameter (step S4). The calculated representative value is accumulated in the representative value data accumulation unit 50 . Representative values are mean, median, mode, and the like. In this embodiment, as an example, an average value is used as a representative value.

Next, based on the master table, the data selection unit 60 selects actual production data in the adoption direction with respect to the average value for each actual production parameter (step S5). Specifically, the data selection unit 60 selects only one of the actual production data smaller than the average value and the actual production data larger than the average value for each actual production parameter.

FIG. 7A is a diagram illustrating a master table. For example, as illustrated in FIG. 7A, multiple types of parameters are set in the master table. As an example, the parameters include MCT average, MCT variation, stage change average, stage change time variation σ, suspension recovery time average, suspension recovery time variation, suspension occurrence frequency, number of resources, work priority consideration, etc. Includes status. MCT average is the average value of MCT. MCT variation is a statistic representing variation such as standard deviation of MCT. The step change average is the average time required for each step change (step change time). The changeover time variation σ is a statistic representing a variation such as the standard deviation of the changeover time. The pause recovery time average is the average value of the time required to recover from the pause. The number of resources is, for example, the number of surplus personnel, such as supporters who perform temporary stoppage recovery work and supporters who perform changeover work.

Improvement directions are defined for each parameter. For example, for the MCT average, the smaller the better the production process. Regarding the number of resources, the more resources, the better the production process. If the direction of improvement is "decrease", the actual production data smaller than the average value will be selected. If the direction of improvement is "increase", the production result data larger than the average value will be selected. As for the MCT average and MCT variation, actual production data smaller than the average value of MCT[min] of the actual production parameter is selected. As for the changeover average and the changeover time variation σ, actual production data that is smaller than the average value of the changeover time [min] of the actual production parameter is selected. Regarding the temporary stop recovery time average and the temporary stop recovery time variation, actual production data that is smaller than the average value for the temporary stop time [min] of the actual production parameter is selected.

As exemplified in FIG. 7(b), production performance data from the minimum value (7 min) to the average value (8 min) are selected for the step change average. As exemplified in FIG. 7(c), production performance data ranging from the minimum value (0 min) to the average value (1.5 min) is selected for the changeover time variation σ.

FIG. 8 is a diagram illustrating actual production data selected for each adjustment parameter. In FIG. 8, the average value of MCT[min] of the actual production parameters is 11.1. Of the actual production data, those whose MCT is less than 11.1 are selected. In the example of FIG. 8, two of production numbers 1 and 3 are selected, and three of production numbers 2, 4 and 5 are not selected. That is, the data to be used is suppressed to 2/5.

Next, the KPI change amount calculation unit 70 calculates the KPI change amount by changing the actual production parameter using the selected actual production data (step S6). Here, the slope (correlation) of the KPI change amount with respect to the change amount of the actual production parameter is calculated. For example, when changing the value of MCT[min] of the actual production parameter, the value of the parameter (MCT average or MCT variation) stored in the master table of FIG. 7(a) may be changed.

The calculated degree of correlation is stored in the degree-of-correlation storage unit 80 . Next, the display device 104 displays the degree of correlation calculated in step S6 (step S7). Through the above processing, it is possible to calculate the degree of correlation between each production performance parameter and the KPI related to the production performance. The display device 104 may display only the production performance parameter with the highest degree of correlation, or may display the ranking of the degree of correlation.

According to the present embodiment, the representative value calculation unit 40 calculates a representative value for each production performance parameter for the accumulated production performance data. Based on the master table, the data selection unit 60 selects only one of the actual production data smaller or larger than the representative value calculated by the representative value calculating unit 40 for each actual production parameter. The KPI variation calculation unit 70 uses the actual production data selected by the data selection unit 60 to calculate the degree of correlation with the KPI related to the actual production for each actual production parameter. With this configuration, it is possible to reduce the number of actual production data for calculating the degree of correlation with KPI. In addition, since it becomes possible to grasp the magnitude of the degree of correlation, it becomes possible to grasp which production performance parameter should be adjusted to significantly improve the KPI. By selecting the actual production data that will improve the production process and using it as an adjustment parameter, it is possible to eliminate the need for digital planning in cases where the production process is worse than the current situation.

In the above example, the representative value calculator 40 functions as an example of a calculator that calculates a representative value for each production performance parameter for the accumulated production performance data. As an example of a selection unit in which the data selection unit 60 selects only one of the actual production data smaller than the representative value calculated by the calculation unit and the actual production data larger than the representative value calculated by the calculation unit for each actual production parameter based on the master table Function. The KPI change amount calculation unit 70 functions as an example of a calculation unit that calculates the degree of correlation with the KPI related to the production performance for each of the production performance parameters using the production performance data selected by the selection unit.

Although the embodiments of the present invention have been described in detail above, the present invention is not limited to such specific embodiments, and various modifications and variations can be made within the scope of the gist of the present invention described in the scope of claims. Change is possible.

10 performance data accumulation unit 20 statistical processing unit 30 statistical data accumulation unit 40 representative value calculation unit 50 representative value data accumulation unit 60 data selection unit 70 KPI change amount calculation unit 80 correlation degree accumulation unit 100 information processing device

Claims (4)

a calculation unit that calculates a representative value for each production performance parameter for the accumulated production performance data;
Based on a master table in which the direction of improving the production process is determined , for each of the production performance parameters, among the production performance data smaller than the representative value calculated by the calculation unit and the production performance data larger than the representative value, the production process is improved. a selection unit that selects only production performance data in the direction of
and a computing unit that computes a degree of correlation with KPIs related to actual production for each of the actual production parameters, using the actual production data selected by the selection unit. 2. The information processing apparatus according to claim 1 , wherein said representative value is an average value. For the accumulated production performance data, the calculation unit calculates a representative value for each production performance parameter,
Based on a master table in which the direction of improving the production process is determined , for each of the production performance parameters, among the production performance data smaller than the representative value calculated by the calculation unit and the production performance data larger than the representative value, the production process is improved. The selection unit selects only the actual production data in the direction of
An information processing method, wherein a calculation unit calculates a degree of correlation with KPIs related to production results for each of the production result parameters, using the production result data selected by the selection unit. to the computer,
a process of calculating a representative value for each production performance parameter for the accumulated production performance data;
Based on a master table in which a direction for improving the production process is determined , the production process is improved in the production result data smaller than the calculated representative value or the production result data larger than the calculated representative value for each production result parameter. a process of selecting only the direction production result data ;
An information processing program, characterized in that, using the selected production performance data, a process of calculating a degree of correlation with KPIs related to the production performance for each of the production performance parameters is executed.

Images