JP2015148961A - Standard working time estimation device and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To estimate a standard working time suited to attributes of work by referring to attribute item values of attribute information related to performed work.SOLUTION: A standard working time estimation device includes: a working type classification section 11 that classifies working time data for each work type item value included in working time history data and generates working time history data; a representative reference time calculation section 12 that calculates a representative reference time for each working type item value from the working time history data; an attribute item classification section 13 that classifies the working time history data for each attribute item value and generates working time history data; a subdivided reference time calculation section 14 that calculates a subdivided reference time for each attribute item value from the working time history data; and a significance evaluation section 15 that sets the subdivided reference time that is determined to have significance, as a reference time of the attribute item value and, for attribute item values other than that, sets the representative reference time as the reference time.

Description

  The present invention relates to a standard work time estimation device and program, and more particularly to a technique for estimating standard work time based on work time required for work performed.

  When planning a maintenance work for facility equipment, for example, an average value or the like is calculated and used as a reference from the work time required for the maintenance work performed in the past. Moreover, when performing work continuously, the technique which sets the work time of a back | latter stage with reference to the work time of the front | latter stage is proposed (for example, patent document 1).

JP 2004-246628 A JP 2009-289134 A JP 2010-128654 A

  It is an object of the present invention to estimate a standard work time suitable for a work attribute by referring to an attribute item value of attribute information related to the work performed.

  The standard work time estimation device according to the present invention includes a plurality of work time data including at least item values indicating types of work performed, work time required for the work, and attribute information related to the work. Work type data for each work type for generating work time data for each work type by classifying each work type item value indicating the type of work included in the time data; and The attribute item value-specific work time data classification means for generating the attribute item value-specific work time data by classifying each attribute item value of the attribute information included in the work type-specific work time data, and the attribute item value-specific work time Corresponding to the work type item value and the attribute item value for each attribute item value for each work type item value included in the work time data by the attribute item value based on the data A subdivision work time calculation means for calculating a work time required for the work, and the work time calculated by the subdivision work time calculation means is a standard required for the work corresponding to the work type item value and the attribute item value. Estimated work time.

  Further, a representative that calculates a standard work time required for work corresponding to the work type item value for each work type item value included in the work type work time data based on the work type work time data. When the standard work time calculated by the work time calculation means and the subdivision work time calculation means is a subdivision work time, and the standard work time calculated by the representative work time calculation means is a representative work time, Significance evaluation means for evaluating the significance of each subdivision work time based on the difference from the representative work time, and if the subordinate work time is evaluated as significant by the significance evaluation means, The standard work time required for work corresponding to the value and the attribute item value is estimated.

  Further, when the significance evaluation means does not evaluate as significant, the representative work time is estimated as the standard work time required for work corresponding to the work type item value and the attribute item.

  Further, the significance evaluation means calculates a p-value by testing a difference between each sub-work time and the representative work time by a test of a population mean, and obtains a minimum value of the p-value for each work type,

  The subdivision work time corresponding to the minimum value of the obtained p value is estimated as the standard work time required for the work corresponding to the work type item value and the attribute item value.

  Further, when the minimum value of the p value obtained by the significance evaluation means is less than a predetermined significance level, the subdivision work time corresponding to the minimum value of the p value that is less than the significance level is set as the work type. The standard work time required for the work corresponding to the item value and the attribute item value is estimated.

  The program according to the present invention allows a computer to store a plurality of work time data including at least item values indicating types of work performed, work time required for the work, and attribute information related to the work. Work type data for each work type for generating work time data classified by work type by classifying each work type item value indicating the type of work included in the data, and further adding the work type data for the work type to the work type The attribute item value-specific work time data classification means for generating the attribute item value-specific work time data by classifying each attribute item value of the attribute information included in the separate work time data, based on the attribute item value-specific work time data For each attribute item value for each task type item value included in the task time data for each attribute item value, it corresponds to the corresponding task type item value and the corresponding attribute item value. A standard work required for the work corresponding to the work type item value and the attribute item value is made to function as a fine work time calculation means for calculating work time required for work, and the work time calculated by the fine work time calculation means Estimate time.

  According to the present invention, it is possible to estimate a standard work time adapted to a work attribute by referring to the attribute item value of the attribute information related to the work performed.

  In addition, only when the work time calculated for each attribute item value for each work type item value is significant, the work time is estimated as the standard work time required for the work corresponding to the work type item value and the attribute item value. Can do.

  Further, only when the work time calculated for each attribute item value for each work type item value is not significant, the representative work time can be estimated as the standard work time required for the work corresponding to the work type item value and the attribute item value. it can.

  Also, if the difference between the subdivision work time and the representative work time is small by performing the test of the population average, the subdivision work time is estimated as the standard work time required for the work corresponding to the work type item value and the attribute item value. can do.

It is the block block diagram which showed the reference time estimation apparatus which is one Embodiment of the standard work time estimation apparatus which concerns on this invention. It is a hardware block diagram of the computer which forms the reference time estimation apparatus in this Embodiment. It is the figure which showed the data structural example of the working time history data accumulate | stored in the working time history data storage part in this Embodiment. It is the flowchart which showed a part of reference time calculation process in this Embodiment. It is the flowchart which showed the processes other than the part shown to FIG. 4A among the calculation processes of the reference time in this Embodiment. It is the figure which showed the data structural example of the work type list | wrist in this Embodiment. It is a figure showing an example of composition of work time history data narrowed down based on work type in this embodiment. It is a figure used in order to demonstrate the calculation method of the reference time in this Embodiment. It is the figure which showed the data structural example of the representative reference time information in this Embodiment. It is the figure which showed the data structural example of the attribute item information in this Embodiment. It is the figure which showed the data structural example of the attribute item value information in this Embodiment. It is a figure showing an example of composition of work time history data narrowed down based on work kind and an attribute item value in this embodiment. It is a figure used in order to demonstrate the calculation method of the reference time for every attribute item value in this Embodiment. It is the figure which showed the data structural example of the subdivision reference time information in this Embodiment. It is the figure which showed the data structural example of the p value information containing the p value calculated for every attribute item value in this Embodiment. It is the figure which showed the data structural example of the p value information containing the minimum value of p value for every work type in this Embodiment. It is the figure which showed the data structural example of the p value information which is significant from the minimum value of the p value for every work type in this Embodiment. It is the figure which showed the data structural example of the reference time information finally obtained in the reference time calculation process in this Embodiment.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram showing a reference time estimation device 10 which is an embodiment of a standard work time estimation device according to the present invention. FIG. 2 is a hardware configuration diagram of a computer forming the reference time estimation device 10 in the present embodiment. The computer forming the reference time estimation device 10 in the present embodiment can be realized by a general-purpose hardware configuration that has existed in the past. That is, the computer is provided as a CPU 21, ROM 22, RAM 23, HDD controller 25 connected with a hard disk drive (HDD) 24, a mouse 26 and a keyboard 27 provided as input means, and a display device as shown in FIG. An input / output controller 29 for connecting each display 28 and a network controller 30 provided as a communication means are connected to an internal bus 31.

  The reference time estimation device 10 in the present embodiment estimates the work time required for maintenance work. In the present embodiment, the estimated work time is particularly referred to as “reference time”. To do. That is, the “reference time” in the present embodiment is to estimate the work time for each work included in the day when planning a maintenance work for a period of a day, for example. This is the standard work time that is used as a reference when estimating the time. In a broad sense, it may be referred to as standard time or standard work time. The work includes types of maintenance, repair, inspection, and the like, but actions that require time other than work actually performed as maintenance work, such as movement, may be added to the work.

  The present embodiment is characterized in that the standard work time to be estimated, that is, the reference time, is not limited to the work type classification level but is further subdivided and estimated for each attribute item value.

  Returning to FIG. 1, the reference time estimation device 10 according to the present exemplary embodiment includes a work type classification unit 11, a representative reference time calculation unit 12, an attribute item classification unit 13, a subdivision reference time calculation unit 14, a significance evaluation unit 15, a reference A time output unit 16 and a work time database (DB) 17 are provided. The work type classifying unit 11 is provided as a work type-specific work time data classification unit, and a plurality of work time history data stored in the work time database 17 is used to indicate a work type included in the work time data. Classification is made for each item value, and work time data by work type ("work time history data D '" described later) is generated. The representative reference time calculation unit 12 is provided as representative work time calculation means, and for each work type item value included in the work type work time data, based on the work type work time data, the work type item. The reference time required for the work corresponding to the value is calculated.

  The attribute item classification unit 13 is provided as an attribute item value-specific work time data classification unit, and further classifies the work type-specific work time data for each attribute item value of the attribute information included in the work type-specific work time data. Work time data by attribute item value ("work time history data D" "described later) is generated. The subdivision reference time calculation unit 14 is provided as subdivision work time calculation means, and based on the work item data by attribute item value, for each work item item value and for each attribute item value included in the work item data by attribute item value. The reference time required for the work corresponding to the work type item value and the attribute item value is calculated.

  The significance evaluation unit 15 is provided as a significance evaluation unit, and each reference time (subdivision work time) calculated by the subdivision reference time calculation unit 14 and the reference time (representative work time) calculated by the representative reference time calculation unit 12. ) The significance of each subdivision work time is evaluated based on the difference from). The reference time output unit 16 outputs the reference information obtained by the reference time estimation device 10.

  FIG. 3 is a diagram showing a data configuration example of work time history data accumulated in the work time database 17 in the present embodiment. The work time history data is composed of at least a set of a work start date and time indicating the execution start date and time of the performed work, a work type indicating the type of the work, a work time indicating the length of time required for the work, and attribute information. The The attribute information includes attribute items such as the target model that indicates the type of equipment that is the target of maintenance work, the branch that indicates the maintenance work location, and the contract information that indicates the details of the contract with the branch. ing.

  Each component 11-16 in the reference time estimation apparatus 10 is implement | achieved by cooperation operation | movement of the computer which forms the reference time estimation apparatus 10, and the program which operate | moves by CPU21 mounted in the computer. The work time database 17 is realized by the HDD 4 mounted in the reference time estimation device 10. Alternatively, the RAM 3 or an external storage means may be used via a network.

  Further, the program used in this embodiment can be provided not only by communication means but also by storing it in a computer-readable recording medium such as a CD-ROM or DVD-ROM. The program provided from the communication means or the recording medium is installed in the computer, and various processes are realized by the CPU of the computer sequentially executing the program.

  Next, reference time calculation processing in the present embodiment will be described with reference to the flowcharts shown in FIGS. 4A and 4B.

  First, the work type classification unit 11 reads all work time history data accumulated in the work time database 17 (step 101). The read work time history data group is represented as work time history data D. Then, the work type classifying unit 11 extracts all work types (work type item values) with reference to the work type items of the work time history data D (step 102). Then, a work ID (Wi (i = 1 to n, n is the number of work types)) is assigned as identification information for identifying the work type for each extracted work type item value, and a work type list is generated. FIG. 5 shows a data configuration example of the generated work type list. After initializing variables i, j, and k used in the processing described later to 1 (step 103), the reference time estimation apparatus 10 repeatedly executes the following processing (steps 104 to 118) for each work type.

  First, the work type classification unit 11 narrows down the work time history data to be processed by extracting all the work time history data corresponding to the work type having the work ID Wi from the work time history data D (step 104). ). The work time history data group narrowed down by this processing is represented as work time history data D ′. FIG. 6 shows an example when work time history data whose work type is maintenance (work ID = W1) is extracted.

  Subsequently, the representative reference time calculation unit 12 refers to the work time item of the extracted work time history data D ′, sets Ri whose work time is equal to or less than Ri, and 90% of the work ID is Wi. Is calculated as a reference time (hereinafter, simply referred to as “work type Wi”) (step 105). A method of calculating the reference time Ri will be specifically described with reference to FIG.

  In FIG. 7, the horizontal axis represents the work time (actual value) required for each work, and the vertical axis represents the number of work (actual number) that required each work time. A value obtained by adding all the numbers of tasks that require each work time corresponds to the total number of work time history data of the work type Wi accumulated in the work time database 17. In FIG. The enclosed range represents 100%. The vertical line (reference time line) 42 indicating the reference time is gradually shifted from the minimum work time to the maximum work time, and is surrounded by a horizontal axis, a graph line 41 indicating the number of work results, and a reference time line 42. When the number of tasks included in the range reaches 90% of the total, the work time indicated by the reference time line 42 is calculated as the reference time of the work type Wi. In the present embodiment, the reference time calculated for each work type is also referred to as “representative reference time”.

  Although the process of step 105 is repeatedly performed, FIG. 8 shows a part after setting the reference time Ri for all work types. As shown in FIG. 8, the representative reference time calculation unit 12 corresponds to the work type Wi included in the work time history data D ′ because it is used to calculate the p value when obtaining the reference time Ri. The number n of records of work time history data and the standard deviation Si of work time are also calculated.

  By the way, although a plurality of items of attribute information are set in the work time history data as shown in FIG. 3, each attribute item has Xj (attribute item ID) as identification information (attribute item ID) for identifying the attribute item. j = 1 to m, where m is the number of attribute items). FIG. 9 shows a data configuration example of attribute item information indicating the correspondence between the attribute item ID and the attribute item. Subsequently, in the reference time estimation device 10, the following processing (steps 106 to 113) is repeatedly executed for each attribute item Xj.

  First, the attribute item classification unit 13 refers to the attribute information of the work time history data D ′ and extracts all the attribute item values existing in the work time history data D ′ for each attribute item Xj (step 106). FIG. 10 shows the attribute item values extracted for the attribute items of the target model (X1) and the branch office (X2), and the attribute item value ID (Xjk (k) as identification information given to the attribute item value. = 1 to Lj and Lj are related to the number of attribute item values in the attribute item Xj)).

  When the attribute item value Xjk is extracted, the reference time estimation device 10 repeatedly executes the following processing (steps 107 to 110) for each attribute item value Xjk.

  First, the attribute item classification unit 13 narrows down the work time history data to be processed by extracting the work time history data whose attribute item Xj is the attribute item value Xjk from the work time history data D ′ (step 107). ). The work time history data group narrowed down by this processing is represented as work time history data D ″. In FIG. 11, the work type is maintenance (work ID = W1), and the target model (attribute item ID = X1). An example when model A (attribute item value ID = X11) is extracted as the attribute item value is shown.

  Subsequently, the subdivision reference time calculation unit 14 refers to each work time of the work time item of the extracted work time history data D ″, and determines Rijk whose work time is equal to or less than Rijk, which is 90% of the total. It is set as a reference time when the work type is Wi and the attribute item Xj is the attribute item value Xjk (step 108) In this process, the reference time calculated by subdividing each attribute item value is further divided into “subdivision”. Also referred to as “reference time”.

  This subdivision reference time Rijk will be described with reference to FIG. FIG. 12 is a graph in which the horizontal axis is the work time and the vertical axis is the number of work, as in FIG. 7. As an example, the work type is maintenance W1 (ie, i = 1) and the attribute item is branch office X2 (ie, j = 2). ), The reference time of branch office a (X21), branch office b (X22), and branch office c (X23) included in the attribute item value of branch office X2 is calculated. The calculation method described with reference to FIG. 7 may be applied as it is as the reference time Rijk calculation method. However, while all the work time history data D is used in FIG. 7, here, the sub-reference time Rijk is calculated using the work time history data D ″ narrowed down by the attribute item value Xjk.

  Until the processing is completed for all the attribute item values Xjk (N in Steps 109 and 110), the processing in Steps 107 and 108 is repeated. FIG. 13 shows all the attribute item values Xjk. A part after the subdivision reference time Rijk is calculated is shown. As shown in FIG. 13, the subdivision reference time calculation unit 14 uses the record of the work time history data included in the work time history data D ″ for use in calculating the p value when obtaining the subdivision reference time Rijk. The number n and the standard deviation Si of the working time are also calculated together.

  When the subdivision reference time is set for each attribute item value (Y in step 110), the significance evaluation unit 15 then calculates the difference between the subdivision reference time Rijk and the representative reference time Ri calculated in step 105 as the population average. The p-value pijk is calculated by the test (step 111).

  In the present embodiment, the p-value is calculated using the t-test as a test for the population mean, but the p-value may be obtained using a method other than the t-test. In the present embodiment, the p-value is calculated by a test method when no equal variance is assumed, and is compared with the significance level as will be described later. Here, the “p value” refers to the probability that a value greater than or equal to | t | occurs in the t distribution of degrees of freedom.

  Until the processing is completed for all the attribute items Xj (N in Steps 112 and 113), the processing in Step 111 is repeated. An example of the result of calculating the p value pijk by this processing is shown in FIG. Shown in

  When the p value pijk is calculated for each work type Wi attribute item value Xjk (Y in step 113), the significance evaluating unit 15 subsequently obtains the minimum value pmin of the p value pijk for each work type Wi. Further, the attribute item corresponding to the minimum value pmin is set as Xisel (step 114). An example of the processing result is shown in FIG. Referring to the numerical examples in FIGS. 14 and 15, for example, when the work type is maintenance (W1), the p value of model B is 0.01, which is extracted as the p value pmin of maintenance (W1). The target model (X1), which is an attribute item corresponding to the model B (X12), is set as Xisel.

  Although the p value (pmin) is obtained for each work type by the above processing, the significance evaluation unit 15 subsequently determines the significance of the obtained p value. In the present embodiment, a numerical value of 0.05 is used as a threshold (significance level) for comparing p values when determining significance, but other values may be used.

  If the p value (pmin) is smaller than 0.05 (Y in step 115), the significance evaluation unit 15 determines that the value is significant, and sets Xisel as a condition item for the work type Wi (step 116). . On the other hand, if the p value (pmin) is 0.05 or more (N in step 115), it is determined that the p value is not significant and the condition item is not set. Insignificant means that even if the reference time is obtained for each attribute item value, the obtained subdivision reference time is not so different from the representative reference time of the work type, in other words, the subdivision reference time corresponding to pmin is not so significant. That is not a good value. An example of the data structure of the p-value information obtained as a result of determining the significance of this pmin is shown in FIG. As can be seen by referring to the numerical examples in FIGS. 15 and 16, since the pmin for repairing the work type W2 is 0.06 and the significance level is 0.05 or more, it is determined that it is not significant.

  The above-described processing (steps 104 to 118) is repeatedly performed for all work types Wi (N in steps 117 and 118). When the processing for all work types Wi is completed, reference time output is performed. The unit 16 generates reference time information with reference to the evaluation result by the significance evaluation unit 15 and outputs it. An example of the data structure of the reference time information finally obtained is shown in FIG.

  As can be understood with reference to FIG. 17, in the present embodiment, the reference time (subdivided) is subdivided for each attribute item rather than for each work type, more strictly, for each attribute item value set in the attribute item. Reference time) is calculated. In particular, in the present embodiment, the sub-reference time determined to be significant as a result of determining the significance of the sub-reference time is set as the reference time, so that a significant reference time can be estimated.

  On the other hand, if the significance of the subdivision reference time is determined to be insignificant, it is determined that the subdivision reference time is not so different from the representative reference time and is not subdivided. In this case, the attribute item value For the above, the representative reference time is set as the reference time without setting the reference time individually. In FIG. 17, when “NULL” is set in the Xisel item, it means that the representative reference time is used for this item.

  The subdivision reference time obtained in step 108 without determining the significance at all may be used as the estimated reference time. However, since it is not necessary to expand records for each work type attribute item value by determining the significance and setting “NULL” in the Xisel item in FIG. 17 and using the representative reference time as the reference time, There is no need to include it in the reference time information shown in FIG. For example, when the work type W2 is repaired, it is obvious that the representative reference time is used without checking the attribute item value.

  The reference time output unit 16 displays the reference time estimated as described above on the display 28, saves it in a file, outputs it from a printer (not shown), or transmits it to another device via a network. Also good. The maintenance work planner makes a maintenance work plan with reference to the reference time.

  In the reference time information finally obtained shown in FIG. 17, only one attribute item value is shown as Xisel, and a reference time is obtained for each attribute item value. In A, the branch office may estimate the reference time corresponding to a combination of a plurality of attribute item values, such as branch office b.

  In the present embodiment, the minimum p value is extracted from the p value information shown in FIG. 14 in step 114, and then compared with the significance level in step 115. If processing is performed in this way, even if the p value is smaller than the significance level, it is not extracted unless it is the minimum value. Therefore, in step 114, each p value is individually compared with the significance level without extracting the minimum value. You may make it determine significance.

  DESCRIPTION OF SYMBOLS 10 Reference time estimation apparatus, 11 Work type classification | category part, 12 Representative reference time calculation part, 13 Attribute item classification | category part, 14 Subdivision reference time calculation part, 15 Significance evaluation part, 16 Reference time output part, 17 Work time database (DB) ), 21 CPU, 22 ROM, 23 RAM, 24 hard disk drive (HDD), 25 HDD controller, 26 mouse, 27 keyboard, 28 display, 29 I / O controller, 30 network controller, 31 internal bus.

Claims (6)

A plurality of work time data including at least item values indicating the type of work performed, the work time required for the work, and attribute information related to the work, and the types of work included in the plurality of work time data Work type data classified by work type for generating work time data classified by work type by classifying each work type item value shown,
The work item data classified by attribute item value for further classifying the work item data classified by work type for each attribute item value of the attribute information included in the work type classified work time data. Means,
Based on the attribute item value work time data, for each work type item value attribute item value included in the attribute item value work time data, the work corresponding to the work type item value and the attribute item value Subdivided work time calculating means for calculating the work time required for
Have
A standard work time estimation device that estimates the work time calculated by the subdivided work time calculation means as a standard work time required for work corresponding to the work type item value and the attribute item value. Based on the work type work time data, for each work type item value included in the work type work time data, the representative work time for calculating the standard work time required for the work corresponding to the work type item value A calculation means;
When the standard work time calculated by the subdivision work time calculation means is subdivision work time and the standard work time calculated by the representative work time calculation means is representative work time, the difference between each subdivision work time and the representative work time A significance evaluation means for evaluating the significance of each subdivision work time based on
Have
The said subdivision work time is estimated as the standard work time required for the operation | work applicable to the said work type item value and the said attribute item value, when it evaluates with the significance by the said significance evaluation means. Standard working time estimation device.   The representative work time is estimated as a standard work time required for work corresponding to the work type item value and the attribute item when the significance evaluation means does not evaluate as significant. Standard work time estimation device. The significance evaluation means calculates the p-value by examining the difference between each sub-work time and the representative work time by the test of the population mean, and obtains the minimum value of the p-value for each work type,
3. The standard according to claim 2, wherein the subdivision work time corresponding to the minimum value of the obtained p value is estimated as a standard work time required for work corresponding to the work type item value and the attribute item value. Work time estimation device.   When the minimum value of the p value obtained by the significance evaluation means is less than a predetermined significance level, the subdivided work time corresponding to the minimum value of the p value that is less than the significance level is set as the work type item value. The standard work time estimation device according to claim 2, wherein the standard work time is estimated for a work corresponding to the attribute item value. Computer
A plurality of work time data including at least item values indicating the type of work performed, the work time required for the work, and attribute information related to the work, and the types of work included in the plurality of work time data Work time data classification means by work type for generating work time data by work type by classifying each work type item value shown,
The work item data classified by attribute item value for further classifying the work item data classified by work type for each attribute item value of the attribute information included in the work type classified work time data. means,
Based on the attribute item value work time data, for each work type item value attribute item value included in the attribute item value work time data, the work corresponding to the work type item value and the attribute item value Sub-work time calculation means for calculating work time required for
Function as
A program that estimates the work time calculated by the subdivided work time calculation means as a standard work time required for work corresponding to the work type item value and the attribute item value.

Images