JP6484536B2 - Work support device and work support method - Google Patents

Description

  The present invention relates to a work support apparatus and a work support method for restoring a manufacturing apparatus or an infrastructure device.

In order to use factory manufacturing equipment, elevators, infrastructure equipment such as railway vehicles (hereinafter referred to as maintenance target devices) at a high operating rate, the maintenance target devices have abnormalities or failures (hereinafter referred to as abnormalities). When this occurs, it is necessary to quickly identify and implement a measure to be performed in order to eliminate the abnormality of the maintenance target device and restore it.
For this reason, a work support system that calculates work to be performed in order to restore the maintenance target device has been proposed.

  For example, Patent Document 1 discloses a maintenance support system that calculates a necessary diagnostic work or a treatment work among a candidate for a treatment work for recovering from an abnormality that has occurred and a candidate for a diagnostic work for specifying a necessary treatment work. Is described. The maintenance support system includes maintenance support tree information in which treatment work candidates and diagnosis work candidates are configured in a tree structure, cost information of each work constituting the maintenance support tree information, the number of past cases, Using the experience value and the information of determination certainty that the result determined as Yes or No in the diagnosis work is the correct probability, the optimum work that minimizes the expected value of work cost or work time is calculated.

Japanese Patent No. 5695998

  In the maintenance support system described in Patent Literature 1, a work order for performing a candidate for a diagnostic work and a candidate for a treatment work is defined in information configured in a tree structure, and an optimal work is simulated according to the work order. To calculate the optimal work. Therefore, in order to calculate the optimum work, it is necessary to define the work order in advance. However, there are cases where the work order cannot be defined in advance, for example, when there are multiple treatment work candidates associated with the result determined as Yes or No in the diagnosis work. Even in this case, the optimum diagnosis work and treatment work are calculated. Work support system is required.

  Accordingly, an object of the present invention is to provide a work support apparatus and a work support method for calculating an optimal diagnosis work or treatment work without using information configured in a tree structure indicating the work order.

  In order to solve the above-described problem, the work support apparatus according to the present invention calculates a work work candidate necessary for recovering a maintenance target apparatus from an abnormality or a diagnosis work candidate for specifying the work work. A list table of abnormality IDs assumed for the maintenance target device, a list of candidate treatment work to be restored for each abnormality ID, and the number of times of restoration for each treatment work, recovery probability, and work time data When a table, a list of diagnosis work candidates for each abnormality ID, a work time for each diagnosis work, a data table of candidate diagnosis results, the abnormality ID, and a candidate for the treatment work for each diagnosis work can be restored A data table of conditional probabilities representing the probability that a diagnostic result of the diagnostic work is obtained, a storage unit for storing, an input of an abnormality ID of the maintenance target device, The candidate for the treatment work and the candidate for the diagnosis work corresponding to the normal ID are read out, and the expected value of the work time from the combination of the work orders in which the treatment work is continued after each diagnosis work is first executed to the minimum The first diagnosis of the work order with the smallest expected work time is calculated by calculating the work order with the smallest expected work time from the combination of the work order and the work order for continuing the treatment work. And an optimum work calculation processing unit that outputs work or treatment work as optimum work.

  Further, as another feature of the present invention, in the work support device, in place of the work time of each work stored in the storage unit, work cost or work man-hour is stored, and the optimum work calculation processing unit is Instead of calculating the expected value of work time until recovery in each work order, calculate the expected value of work cost until recovery in each work order, or the expected value of work man-hours until recovery in each work order To be configured.

  According to still another aspect of the present invention, the work support device further includes a work support main processing unit, and the work support main processing unit is calculated by the optimum work calculation processing unit and presented to the operator. The operator accepts the result of execution of the optimum work by the worker, and if the abnormality of the maintenance target device has not been resolved, the treatment work corresponding to the abnormality ID stored in the storage unit The candidate recovery probability is updated, and the optimum work calculation processing unit is activated again to calculate the next optimum work.

  Further, in order to solve the above-described problem, the work support method of the present invention uses information processing by a computer to specify a candidate treatment work or a treatment work necessary for recovering a maintenance target device from an abnormality. A work support method for calculating diagnostic work candidates, comprising: a list table of abnormality IDs assumed for the maintenance target device; a list of treatment work candidates for recovery for each abnormality ID; Data table of restoration frequency, restoration probability, work time, list of candidates for diagnosis work for each abnormality ID, work time for each diagnosis work, data table of candidate for diagnosis results, abnormality ID, and each diagnosis work A conditional probability data table representing the probability of obtaining a diagnostic result of the diagnostic work when it can be recovered with the candidate treatment work, A combination of work orders that accepts an input of an abnormality ID of a maintenance target device, reads out the candidate for the treatment work and the candidate for the diagnosis work corresponding to the abnormality ID, and continues the treatment work after each diagnosis work is first executed Calculating the work sequence with the smallest expected value of the work time from recovery to the recovery and the work order with the minimum expected value of the work time until recovery from the combination of the work orders in which the treatment work is continued. The first diagnosis work or the work work in the work order in which the expected value of the work time is small is configured to be presented as the optimum work.

  According to the present invention, a candidate for a treatment work and a diagnostic work are created by creating a work order for a candidate for a treatment work and a candidate for a diagnostic work by a work order creation process and specifying an optimal work order by the work order specification process Even when the candidate work order is not defined in advance, the work required for quick recovery can be calculated.

It is a flowchart which shows the work assistance procedure when utilizing the work assistance apparatus which concerns on one embodiment of this invention. It is a block diagram which shows the structure of the outline of the work assistance apparatus which concerns on one embodiment of this invention. It is a table | surface which shows the table example of the abnormality list data based on one embodiment of this invention. It is a table | surface which shows the table example of the treatment work data which concerns on one embodiment of this invention. It is a table | surface which shows the table example of the diagnostic work data which concerns on one embodiment of this invention. It is a table | surface which shows the example of a table of conditional probability data which concern on one embodiment of this invention. It is a flowchart which shows the optimal work calculation process flow of the optimal work calculation process part which concerns on one embodiment of this invention. It is a detailed process flow figure of S302 of the optimal work calculation process which concerns on one embodiment of this invention. It is a detailed process flow figure of S304 of the optimal work calculation process which concerns on one embodiment of this invention. It is a table | surface which shows the example of a table of the work order data which concerns on one embodiment of this invention. It is a table | surface which shows the table example of the expected value calculation result data which concerns on one embodiment of this invention. It is a table | surface explaining the process of the expected value calculation which concerns on one embodiment of this invention. It is a table | surface which shows the example of a table of treatment work order data which concerns on one embodiment of this invention. It is a figure which shows the display screen which concerns on one embodiment of this invention. It is a figure which shows the display screen which concerns on one embodiment of this invention. It is a figure which shows the display screen which concerns on one embodiment of this invention. It is a block diagram which shows the structure of the outline of the work assistance system which concerns on 2nd embodiment of this invention.

The present invention relates to a work support system that calculates work necessary for resolving an abnormality occurring in a device such as a manufacturing apparatus or an infrastructure device, and is a candidate for a treatment work for resolving the abnormality. Then, a work order is created among the candidates for the diagnosis work for specifying the necessary treatment work, and the necessary work is calculated using the expected value of the work time or work cost calculated based on the work order.
Hereinafter, embodiments of the present invention will be described with reference to the drawings. Note that components having the same function are denoted by the same reference symbols throughout the drawings for describing the embodiment, and the repetitive description thereof will be omitted.

  FIG. 1 is a flowchart showing a work procedure when an administrator of a maintenance target device or an operator such as a maintenance staff uses the work support device 100 of the present invention to solve an abnormality of the maintenance target device and restore it. . In this example, we will explain the work to solve and restore the abnormality of the maintenance target device, but we will resolve the failure of the maintenance target device, eliminate the sign of abnormality, eliminate the failure of the maintenance target device, etc. The same applies to the above. Also, the resolution and restoration of abnormality are used in the same meaning.

  When an administrator of the maintenance target device or an operator such as a maintenance staff inputs an execution instruction of the optimal work calculation process to the work support device 100 in order to eliminate the abnormality of the maintenance target device (S101), the work support device 100 is optimal. An optimal work calculation process for calculating an appropriate work and presenting it to the worker is performed (S201). Details of this process S201 are shown in an optimal work calculation process flow of FIG. 7 and detailed process flows of FIGS.

  The worker executes the calculated diagnostic work or treatment work (S102), and inputs the result after the work is executed (S103).

  When the operator performs the treatment work and the abnormality is resolved, the process proceeds to an end instruction step (S104). If the abnormality has not been eliminated, that is, if the diagnosis work has been executed, or if the abnormality has not been eliminated by performing the treatment work, the process returns to the execution of the work (S102). If the abnormality has not been resolved, the work support apparatus 100 proceeds to the update process (S203) of the recovery probability of the treatment work held in the storage unit based on the input of the work result from the worker (S103) (S202). ). After the probability update process (S203), the process returns to the optimum work calculation process (S201), and the optimum work is calculated with the updated probability.

The operator ends the work when the abnormality is resolved (S105). In addition, the work support apparatus 100 performs processing for updating the treatment work data. The conditional probability data is updated if necessary, but the principle is not updated (S204). The data updated here is used for optimal work calculation processing when a new abnormality occurs next time.
<Work support device 100>
FIG. 2 is a block diagram illustrating a configuration of the work support apparatus 100. The work support apparatus 100 can be configured on a general-purpose computer, and the hardware configuration thereof includes a calculation unit 110 including a CPU (Central Processing Unit), a RAM (Random Access Memory), and a ROM (Read Only). Memory 120, HDD (Hard Disk Drive), storage unit 120 composed of SSD (Solid State Drive) using flash memory, etc., input unit 130 composed of input devices such as keyboard and mouse, CRT display, LCD (Liquid Crystal Display), a display device 140 such as an organic EL display, a display unit 140 constituted by various output devices, and a media reading unit 150 that reads information from a portable storage medium having portability such as a CD-ROM and a USB memory , And a communication unit 160 constituted by a NIC (Network Interface Card) or the like.

  The communication unit 160 is connected via a network 170 to a maintenance target device 180 to be diagnosed, a portable terminal 190 carried by an operator, and the like. The network 170 is a communication network such as the Internet or an intranet.

  The calculation unit 110 implements the following functional units by loading the work support program 121 stored in the storage unit 120 into the RAM and executing it by the CPU. The calculation unit 110 includes a work support main processing unit 111, a database registration processing unit 112, and an optimum work calculation processing unit 113.

  The storage unit 120 includes a work support program storage area 121, an abnormality list data storage area 122 that stores a list of abnormalities that occur in the maintenance target device 180, a list of treatment work candidates, the number of times each work was restored, A treatment work data storage area 123 that stores the probability of recovery by the treatment work (probability of eliminating the abnormality) and the work time of the treatment work, and a diagnosis work data storage that stores a list of diagnosis work candidates and the work time of the diagnosis work Area 124, conditional probability data storage area 125 for storing the probability of obtaining the result of the diagnostic work when each treatment work is restored, and the work of the diagnostic work or the treatment work created in the optimum work calculation process described later A work order data storage area 126 for storing the order, and an expected value meter for storing the expected value for each work calculated in the optimum work calculation process described later A result data storage area 127, and a treatment work order data storage area 128 for storing the working order of the treatment operations only created at the optimum working calculation process to be described later, the.

The work support main processing unit 111 of the calculation unit 110 controls the entire work support process (S201 to S204 in FIG. 1).
Prior to the work support process, the database registration processing unit 112 stores each data serving as a database in the abnormality list data storage area 122, the treatment work data storage area 123, the diagnostic work data storage area 124, and the conditional probability data storage area 125. sign up. The registration process is performed by reading from a portable storage medium that records data created elsewhere by a user input from the input unit 130, or data created by another system via the network 170. The data is received and stored in the storage unit 120.
The optimal work calculation processing unit 113 includes a recovery probability and work time stored in the treatment work data storage area 123 of the storage unit 120, a work time stored in the diagnostic work data storage area 124, and a conditional probability data storage area 125. The process of calculating the optimum work is performed using the conditional probability data stored in. Details of the optimum work calculation processing performed by the optimum work calculation processing unit 113 will be described later.

The maintenance target device 180 is a maintenance target device in which an abnormality has occurred in this embodiment. A configuration may be adopted in which a plurality of maintenance target devices are connected.
Further, the maintenance target device 180 is not necessarily connected to the work support device 100 via the network 170, and may be configured independently.
<Storage unit 120>
In the storage unit 120, storage areas of an abnormality list data storage area 122, a treatment work data storage area 123, a diagnostic work data storage area 124, and a conditional probability data storage area 125 are set in advance.
<Abnormality list data storage area 122>
In the abnormality list data storage area 122, names of abnormalities that may occur in the maintenance target apparatus 180 or the same type of apparatus as the maintenance target apparatus 180, and alarms that are generated by the maintenance target apparatus 180 when the abnormality occurs are stored. Information for identifying the code is stored.

FIG. 3 shows an abnormality list data table 122 a stored in the abnormality list data storage area 122. This table has fields such as an ID0 column 122b, an abnormal name column 122c, and an alarm code column 122d.
The ID0 column 122b stores information for identifying an abnormality that may occur in the maintenance target device 180 or the same type of device as the maintenance target device 180. For example, natural numbers 1, 2, 3,... Continuous values are stored in order from the top row of the table.
The abnormality name column 122c stores information that identifies the name of an abnormality that may occur in the maintenance target device 180 or the same type of device as the maintenance target device 180.
In the alarm code column 221d, the maintenance target device 180 or the device of the same type as the maintenance target device 180 has an alarm code that is issued by the maintenance target device 180 when an abnormality identified by the ID or abnormality name of the row occurs. Information for identifying is stored.
<Treatment work data storage area 123>
In the treatment work data storage area 123, when an abnormality identified by the information stored in the abnormality list data storage area 122 occurs in the maintenance target device 180, a measure for eliminating the abnormality and recovering from the abnormality Information for specifying work candidates, the past number of times recovered by the treatment work, the probability that each treatment work candidate can recover from the abnormality, and the work time required for each treatment work candidate are stored.

  FIG. 4 shows an example of the treatment work data table 123 a stored in the treatment work data storage area 123. This table has fields such as an ID0 column 123b, an ID1 column 123c, a treatment work content column 123d, a recovery count column 123e, a recovery probability column 123f, and a treatment work time column 123g.

  The ID0 column 123b stores information for specifying the abnormality specified in the abnormality list data table 122a. The treatment work data table 123a is composed of a plurality of tables having different information stored in the ID0 column 123b, and each table is for recovering from an abnormality identified by the information stored in the ID0 column 123b. The information on the candidate treatment work is stored.

The ID1 column 123c stores information for specifying treatment work candidates. For example, values consecutive to A1, A2, A3,... ANx are stored in order from the top row of the table.
The action work content field 123d stores information for specifying the content of the action work for recovering from the abnormality. The recovery number field 123e is the action work specified by the information stored in the ID1 field 123c. Information specifying the number of times of recovery from the abnormality specified by the information stored in the ID0 column 123b is stored. An initial value is stored in advance in the restoration count column 123e by the system designer. The initial value is updated in an optimal work calculation process described later.

  The recovery probability column 123f stores information for specifying a probability that the recovery work can be recovered from the abnormality specified by the information stored in the ID0 column 123b by the treatment work specified by the information stored in the ID1 column 123c. The information stored in the recovery probability column 123f has an initial value stored in advance by the system designer. In addition, it is updated with the information stored in the number-of-recoveries column 123e in the optimum work calculation process described later.

The treatment work time column 123g stores information for specifying the work time required for the treatment work specified by the information stored in the ID1 column 123c. Note that the stored information may be cost information including man-hours required for work and parts replacement costs. The initial value is stored in advance in the action work time field 123g by the system designer, but the work time may be calculated from the operation log of the system and stored.
<Diagnostic work data storage area 124>
In the diagnostic work data storage area 124, when an abnormality identified by the information stored in the abnormality list data storage area 122 occurs in the maintenance target device 180, a measure for eliminating the abnormality and recovering from the abnormality Information for specifying a diagnostic work candidate for specifying a work, a work time for each of the diagnostic work candidates, and a result obtained when the diagnostic work is performed on the diagnostic work candidates is stored.

  FIG. 5 shows a diagnostic work data table 124 a which is an example of the diagnostic work data storage area 124. This table has fields such as an ID0 column 124b, an ID2 column 124c, a diagnostic work content column 124d, a diagnostic work time column 124e, and a diagnostic result column 124f.

  The ID0 column 124b stores information for specifying the abnormality specified in the abnormality list data table 122a. The diagnostic work data table 124a is composed of a plurality of tables in which different information is stored in the ID0 column 124b, and each table can recover from an abnormality identified by the information stored in the ID0 column 124b. Information on candidates for diagnosis work for specifying work candidates is stored.

The ID2 column 124c stores information for specifying diagnostic work candidates. For example, values consecutive to D1, D2, D3,... DNy are sequentially stored from the top row of the table.
The diagnostic work content column 124d stores information for specifying the content of the diagnostic work specified by the information stored in the ID2 column 124c. The diagnostic work time 124e is specified by the information stored in the ID2 column 124c. Information for identifying the work time required for the diagnostic work is stored. The stored information may be cost information including the number of man-hours required for the work and the arrangement cost of equipment. The initial value is stored in advance in the diagnostic work time column 124e by the system designer, but the work time may be calculated from the operation log of the system and stored.

The diagnosis result column 124f stores information for specifying candidates for the result of the diagnosis work specified by the information stored in the ID2 column 124c. In the diagnosis result column 124f, the diagnosis result ID (value before: in FIG. 5) and the contents of the diagnosis result are stored together so that the contents of the diagnosis result do not overlap in each column.
<Conditional probability data storage area 125>
The conditional probability data storage area 125 stores information for specifying the probability that a diagnostic result is obtained for each candidate for the diagnostic work.

FIG. 6 shows a conditional probability data table 125 a which is an example of the conditional probability data storage area 125. This table has fields such as an ID0 column 125b, an ID2 column 125c, a diagnosis result column 125d, and a conditional probability column 125e.
The ID0 column 125b stores information for specifying the abnormality specified in the abnormality list data table 122a.
The ID2 column 125c stores information for specifying the diagnostic work specified in the diagnostic work data table 124a.

  The conditional probability data table 224a is configured by a plurality of table groups having different information stored in the ID0 column 125b, and each table group is configured by a plurality of tables having different information stored in the ID2 column 125c. Has been. Each table stores information for specifying the probability of obtaining each diagnosis result of the diagnosis work specified by the information stored in the ID2 column 125c.

  The diagnosis result column 125d stores information for specifying the result of performing the diagnosis work specified by the information stored in the ID2 column 125c. Therefore, the same information as the information stored in the line specified by the information stored in the ID2 column 125c among the information stored in the diagnosis result column 124f of the diagnostic work data table 124a is stored.

  In the conditional probability column 125e, information for specifying the probability of obtaining each diagnosis result when the candidate for the treatment work stored in the treatment work data table 123a can be recovered from the abnormality is stored. Each column of the conditional probability column 125e is associated with information stored in the ID1 column 123c of the treatment work data table 123a, and is specified by the information stored in the ID1 column 123c of the treatment work data table 123a. The probability of obtaining a diagnostic result when the work can be recovered is shown. Further, each row of the conditional probability column 125e stores information for specifying a probability that the diagnosis result specified in the diagnosis result column 125d is obtained when it can be recovered by the corresponding treatment work in each column. For example, by collecting a plurality of past cases in which anomalies have been recovered in the treatment work A2, and summing up the number of diagnosis results of the diagnosis work D1 in the past cases, R1: the ratio of the number of cases within the normal range to the total number 0.9, R2: The same proportion of cases that are above the normal range is 0.05, and R3: The same proportion of cases that are below the normal range is 0.05.

  For example, in the example of the data table of FIGS. 3, 4, 5, and 6, when an abnormality occurs in which the hot air with ID0 = 1 does not occur, the treatment work of replacing the thermostat with ID1 = A1 is performed. If it is possible to recover from the abnormality (when the thermostat is the cause of failure), the diagnostic work of confirming the operating temperature of ID2 = D1 indicates a relationship of R3: below the normal range with a probability of 0.8. Yes.

Each column of the conditional probability column 125e does not need to include all treatment work candidates stored in the ID1 column 123c of the treatment work data table 123a. Although the column “Others” is stored in the ID1 column 123c of the treatment work data table 123a, a diagnosis result is obtained when the candidate can be recovered with a treatment work candidate not defined as a column in the conditional probability column 125e. Shows the probability of being.
An initial value is stored in advance in the conditional probability column 125e by the system designer. The initial value may be updated in an optimal work calculation process described later.

In the storage unit 120, the data created by the optimum work calculation process shown in S <b> 201 described above are stored in the work order data storage area 126, the expected value calculation result data storage area 127, and the treatment work order data storage area 128. Is set as a work area for temporarily storing.
<Work order data storage area 126>
FIG. 10 shows an example of a work order data table 126 a stored in the work order data storage area 126. The work order data table 126a includes fields such as an ID2 field 126b, an ID3 field 126c, a work order field 126d, an expected value field 126e for each diagnosis result, and an expected value minimum field 126f for each diagnosis result.

The ID2 column 126b stores information for specifying the diagnostic work specified in the diagnostic work data table 124a.
The ID3 column 126c stores information for specifying the diagnosis result. The work order data table 126a is composed of a plurality of tables having different information stored in the ID3 column 126c. Each table includes information indicating the work order of a candidate for a diagnostic work or a candidate for a treatment work, Information indicating an expected value in each work order and information indicating a work order that minimizes the expected value are stored.

The work order column 126d stores the work order of candidates for diagnostic work or candidates for treatment work.
In the expected value column 126e for each diagnosis result, information for specifying the expected value of the work time of each work order is stored.
In the minimum expected value column 126f for each diagnosis result, information for specifying a work order that minimizes the expected value of work time is stored.
<Expected value calculation result data storage area 127>
FIG. 11 shows an example of the expected value calculation result data table 127 a stored in the expected value calculation result data storage area 127. The expected value calculation result data table 127a includes fields such as a start work column 127b, an expected value column 127c for the optimum work order, and a minimum expected value column 127d.

The start work column 127b stores information for identifying a candidate for a diagnostic work or a candidate for a treatment work to be presented to the worker.
The expected value column 127c of the optimum work order shows the work time when the work is performed in the optimum work order in which the expected value of the work time is minimized from the diagnosis work or the treatment work specified in the start work field 127b. Stores information specifying the expected value.
In the minimum expected value column 127d, the start work in which the expected value of the work time stored in the expected value field 127c of the optimal work order is the minimum among the diagnosis work or the treatment work specified in the start work field 127b is specified. Information is stored.
<Procedure work order data storage area 128>
FIG. 13 shows a treatment work order data table 128 a stored in the treatment work order data storage area 128. The treatment work order data table 128a has fields such as a work order field 128b, an expected work time value field 128c, and a minimum expected value field 128d.

The work order column 128b stores work orders of candidate treatment work.
The expected work time value column 128c stores information for specifying the expected work time value for each work order.
The expected value minimum column 128d stores information for specifying a work order that minimizes the expected work time value.
<Optimum Work Calculation Processing of Optimal Work Calculation Processing Unit 113>
As shown in FIG. 1, when the optimum work calculation process is instructed by the worker (S101), the work support apparatus 100 activates the work support main processing unit 111. In the processing flow of the work support main processing unit 111, the optimum work calculation processing unit 113 is activated, and the optimum work calculation processing unit 113 performs optimum work calculation processing for calculating work effective for recovery from an abnormality (S201). ).

  Optimal work calculation processing in this embodiment is optimal for recovery from abnormalities by creating work orders for diagnostic work candidates and treatment work candidates and identifying the work order that minimizes the expected work time. It is characterized by calculating a simple work.

  Hereinafter, the optimum work calculation process will be described with reference to the flowchart of FIG. 7 showing the optimum work calculation process flow. Further, in the description, a flowchart showing the detailed processing flow of FIG. 8 which is the details of the processing of S302 in the flowchart of FIG. 7 and the detailed processing flow of FIG. 9 which is the details of the processing of S304 of the flowchart of FIG. The flowchart shown is used together.

  In the optimum work calculation process, when an abnormality ID that has occurred in the maintenance target device 180 is identified (S300), among the candidate treatment work and the diagnosis work candidate necessary for recovery of the abnormality ID, first, the diagnosis work candidate is selected. The optimum work order is specified, and an expected value of work time is calculated (S302). When the optimum work order is specified for all the candidates for the diagnosis work and the expected work time is calculated (S301 to S303), the expected work time when performing the treatment work without performing the diagnosis work is calculated (S304). ), The diagnostic work or the treatment work included in the work order in which the calculated expected value is the minimum is presented to the worker as the optimum work (S305).

  Next, each process from S300 to S305 will be specifically described.

In step S300, the optimum work calculation processing unit 113 executes the optimum work calculation process when an abnormality that has occurred is selected by the selection button D13 on the input screen of FIG. 14 and the button of the optimum work calculation D14 is pressed. .
FIG. 14, which is an example of an input screen for instructing the execution of the optimum work calculation process by the worker, will be described.
D11 is information for identifying an ID of an abnormality that occurs in the maintenance target device 180, and information stored in the ID0 column 122b of the abnormality list data table 122a is displayed.
D12 is information for identifying the name of an abnormality that occurs in the maintenance target apparatus 180, and information stored in the abnormality name column 122c of the abnormality list data table 122a is displayed.
D13 is a button for selecting an abnormality that has occurred.
D14 is a button for instructing optimal work calculation.

  The selection button D13 may be configured to automatically select using an alarm code generated by the maintenance target device 180. Further, when the maintenance target device 180 issues an alarm code, the optimum work calculation process may be automatically executed.

  The optimum work calculation processing unit 113 reads the abnormality ID (D11) specified by the selection button D13, and performs the processing after S301. Thereafter, the treatment work data table 123a, the diagnostic work data table 124a, and the conditional probability data table 125a are stored in the abnormality ID (D11) specified by the selection button D13 and the respective ID0 columns (123b, 124b, 125b). A data table that matches the information obtained is used.

  In step S301, the optimum work calculation processing unit 113 specifies the optimum work order for the diagnostic work stored in each row of the diagnostic work data table 124a, and the expected value of the work time when the work is performed in the optimum work order. Is calculated. Here, i (i = 1, 2,... Ny) is selected as a row number, the first row to the last row are selected in order, and the processing of S302 up to S303 is repeated.

Details of the processing of S302 will be described using the flowchart of detailed processing of FIG.
In step S400, the optimum work calculation processing unit 113 acquires information stored in the diagnosis result column 124f for the diagnosis work in the i-th row of the diagnosis work data table 124a.

  In S401 to S405, an expected value is calculated for each diagnosis result stored in each column of the diagnosis result column 124f. Here, j (j = 1, 2,..., Nz) is selected as a column number in order from the first column to the last column, and the processing from S401 to S405 is repeated.

  In step S <b> 402, the optimum work calculation processing unit 113 diagnoses the i-th diagnosis work as the first work to calculate the optimum work order and expected work time values for the i-th diagnosis work. The work order of the work and the treatment work is created and stored in the work order column 126d of the work order data table 126a.

  In this embodiment, the i-th diagnosis work is the first work, and all combinations of possible work orders are created for all treatment work candidates stored in the treatment work data table 123a. Store in column 126d. In addition, you may create the work order which has a diagnostic work in the 2nd work after. Further, the combination of work orders may be limited to some treatment work.

  In step S403, the optimum work calculation processing unit 113 calculates the expected value of the work time according to the following (Equation 1) for each work order stored in each row of the work order column 126d of the work order data table 126a. Stored in the expected value column 126e for each diagnosis result of the order data table 126a.

Here, k: k is a natural number taking k = 1, 2,... N, and is a value indicating the line number of the work order column 126d. N indicates the number of rows.
ET _k : A value indicating the expected value of the work time in the work order of the k-th row in the work order column 126d.
m: m = 2, 3,... M is a natural number, and is a value indicating the column of the work order column 126d. M indicates the number of columns.
p (A _km ): A value indicating the recovery probability of the treatment work in the k-th row and m-th column of the work order column 126d. The recovery probability is information specified in the recovery probability column 123f of the treatment work data table 123a.
p (D _i = R _j | A _km ): Conditional probability that the j-th diagnosis result can be obtained in the i-th diagnosis work when recovery can be made by the treatment operation in the k-th row and m-th column of the work order column 126d. is there. Conditional probability is stored in the table ID2 field 125c of the conditional probability data table 125a is _{D i,} diagnosis column 125d is _{R j} of the conditional probability column 125e, treatment work in rows and columns as the _{A miles} Information. In addition, when there is no treatment work corresponding to the column in the conditional probability column 125e, the information stored in the “Others” column is used.
TD _i is a value indicating the diagnostic work time of the i-th diagnostic work. The diagnostic work time is information specified in the diagnostic work time column 124e of the diagnostic work data table 124a.
n: A natural number taking n = 2, 3,... m, and a value indicating the column of the work order column 126d.
TA _kn : a value indicating the treatment work time of the treatment work in the k-th row and the n-th column of the work order column 126d. The treatment work time is information specified in the treatment work time column 123g of the treatment work data table 123a.

The calculation process of the expected value calculation of (Equation 1) is shown in FIG.
d01 indicates a list of the work to be restored and the work performed up to the restoration for the work order of the k-th line (the work order of the first line in FIG. 10 in the example of FIG. 12) that is the target of the expected value calculation. ing.
In the work sequence of the first row in FIG. 10, only the diagnostic work D1 and the treatment work A1 are performed when the treatment work A1 is restored, but the diagnosis work D1, the treatment work A1, and the treatment are performed when the treatment work A2 is restored. Work A2 will be performed. In this way, for each of the treatment work candidates in the work order, the diagnosis work candidates and the treatment work candidates that have been performed before the restoration are shown.

d02 indicates the probability of recovery by each treatment operation corrected by the conditional probability, and represents p (A _km ) × p (D _i = R _j | A _km ) in (Equation 1).
d03 indicates the sum of the work time of the diagnosis work candidate and the treatment work candidate until the recovery work is performed, that is, the sum of the work time required for the recovery, and TD _i + Σ of (Equation 1) D04 representing (TA _kn ) represents the product of the probability of d02 and the work time of d03.
The first summation symbol in (Equation 1) indicates that the sum is taken for each row of d04 as shown in d05, and the value ET _k calculated in (Equation 1) is the probability of d02 and the working time of d04. Represents the sum of products.
When the probability d02 that can be restored when the work time d03 is small, the value of ET _k is small. A smaller ET _k indicates an optimum work order that can be recovered in a shorter work time.

In step S404, the optimum work calculation processing unit 113 compares the information stored in the expected value column 126e for each diagnosis result of the work order data table 126a, specifies the work order having the minimum expected value, and executes the row of the work order. The symbol “◯” indicating the minimum expected value is stored in the minimum expected value column 126f for each diagnosis result.
The work order specified here is the i-th diagnosis work (diagnostic work specified in the ID2 column 126b), and when the j-th diagnosis result (diagnosis result specified in the ID3 column 126c column) is obtained, This is an optimal work order that minimizes the expected value of work time.

  In step S405, the optimum work calculation processing unit 113 calculates the expected value for each diagnosis result for all the diagnosis results of the i-th diagnosis work, and if the optimum work order for each diagnosis result is specified, S406 Proceed to the process. If the expected value for each diagnostic result is not calculated for all the diagnostic results, 1 is added to j and the processing returns to S401, and the expected value for each diagnostic result is calculated for the next diagnostic result.

In step S406, the optimum work calculation processing unit 113 calculates the sum of the minimum expected values for each diagnosis result, starts the i-th diagnosis work, and performs work in the optimum work order for each diagnosis result. The expected value of the working time is calculated.
Specifically, the optimum work calculation processing unit 113 first stores the ID of the i-th diagnosis work in the start work column 127b of the expected value calculation result data table 127a.
Then, a line in which “◯” indicating the minimum value is stored in the minimum expected value column 126f for each diagnosis result of the work order data table 126a is specified, and the information stored in the expected value 126e for each diagnosis result of that line is stored. The data is extracted from each table of the work order data table 126a and stored in the row of the expected value column 127c of the optimum work order in the expected value calculation result data table 127a. Thus, the expected work time value is stored when the work is started from the i-th diagnosis work and the work is performed in an optimal work order that minimizes the expected work time value according to the diagnosis result.
When the process of S406 ends, the optimal work calculation processing unit 113 ends the detailed process flow of FIG. 8 and returns to the flow of the optimal work calculation process of FIG.

  In step S303, the optimum work calculation processing unit 113 performs the process of S302 for all the diagnostic works stored in the diagnostic work data table 124a, and sets values in the expected value column 127c of the optimum work order in the expected value calculation result data table 127a. Is stored, the process proceeds to S304. When the process of S302 is not performed for all the diagnostic work, 1 is added to i and the process returns to the process of S301, and the expected value of the work time in the optimum work order is calculated for the next diagnostic work.

In step S304, the optimal work calculation processing unit 113 calculates an expected value of the work time when the work is performed in the optimal work order for the treatment work candidates stored in each row of the treatment work data table 123a.
Details of the process of S304 will be described using the flowchart of the detailed process of FIG.

In step S500, the optimum work calculation processing unit 113 creates a combination of work orders that can be taken for the candidate treatment work stored in each row of the treatment work data table 123a, and the work order that minimizes the expected value of the work time. Is identified.
In the present embodiment, all combinations of possible work orders are created for all treatment work candidates stored in the treatment work data table 123a and stored in the work order column 128b. Note that the combination of work orders may be limited to some treatment work (for example, treatment work whose treatment work recovery probability 123f is equal to or higher than a predetermined threshold).

  In step S501, for each work order stored in each row of the work order column 128b of the treatment work order data table 128a, an expected value of work time is calculated according to the following (Equation 2), and work in the treatment work order data table 128a is performed. Stored in the expected time value column 128c.

Here, k: k is a natural number taking k = 1, 2,... N, and is a value indicating the line number of the work order column 128b. N indicates the number of rows.
ET _k : A value indicating the expected value of the work time in the work order of the k-th row in the work order column 128b.
m: A natural number taking m = 1, 2,... M, and a value indicating the column of the work order column 128b. M indicates the number of columns.
p (A _km ): a value indicating the recovery probability of the treatment work in the k-th row and m-th column of the work order column 128b. The recovery probability is information specified in the recovery probability column 123f of the treatment work data table 123a.
n: A natural number taking n = 1, 2,..., m, and a value indicating the column of the work order column 128b.
TA _kn : a value indicating the treatment work time of the treatment work in the k-th row and the n-th column of the work order column 128b. The treatment work time is information specified in the treatment work time column 123g of the treatment work data table 123a.

In step S502, the information stored in the expected value column 128c of the work time in the treatment work order data table 128a is compared, the work order with the minimum expected value is specified, and the expected value minimum field 128d in the row of the work order is The symbol “◯” indicating the minimum expected value is stored. In addition, the ID of the first treatment work to be performed in the work order having the minimum expected value is added to an arbitrary line in the start work column 127b of the expected value calculation result data table 127a, and the expected value is the expected value of the optimum work order. It adds to the same line of the column 127c.
As a result, when the treatment work is performed without performing the diagnosis work, the treatment work that minimizes the expected value of the work time is identified.

  When the process of S502 ends, the optimal work calculation processing unit 113 ends the detailed process flow of FIG. 9 and returns to the flow of the optimal work calculation process of FIG.

In step S305, the optimum work calculation processing unit 113 presents the start work that minimizes the expected work time value to the worker as the optimum work.
The optimum work calculation processing unit 113 compares the information stored in the expected value column 127c of the optimum work order in the expected value calculation result data table 127a, identifies the row where the smallest information is stored, and the expected value minimum field 127d. “○” indicating the minimum expected value is stored in the corresponding row.
Further, the optimum work calculation processing unit 113 determines the information stored in the row with the minimum expected value in the start work column 127b, that is, the first diagnostic work or treatment work in the work order in which the expected value of the work time is minimized. To present.

FIG. 15 shows an example of the output screen of the optimum work calculation result displayed on the display screen of the output unit 140 or the display screen of the portable terminal 190 carried by the worker to the installation location of the maintenance target device 180.
In D21, the first diagnosis work or treatment work in the work order in which the expected work time value is calculated to be the minimum is displayed. That is, the ID of the start work (diagnostic work or treatment work) in the line in which “○” indicating the minimum expected value is stored in the expected value minimum column 127d of the expected value calculation result data table 127a, and the diagnostic work associated with the ID The information stored in the content column 124d or the information stored in the treatment work content column 123d is displayed.
The values stored in the expected value column 127c of the optimum work order in the expected value calculation result data table 127a may be ranked, and a plurality of diagnostic work or treatment work may be displayed in order from the lowest order.

  In D22, an expected value of the work time when the work is performed from the diagnosis work or the treatment work displayed in D21 is displayed. That is, the information stored in the expected value column 127c of the optimum work order in the row in which “O” indicating the minimum expected value is stored in the expected value minimum column 127d of the expected value calculation result data table 127a is displayed.

In D23, information indicating the execution result of the diagnostic work or the treatment work is displayed. When the work displayed in D21 is a diagnostic work, the information stored in the diagnostic result column 124f of the diagnostic work data table 124a is displayed. When the work displayed in D21 is a treatment work, information specifying whether or not the abnormality has been resolved is displayed.
D24 is a button for selecting a work result.
D25 is a button for inputting the result of the work selected in D24.

  In D26, a graph model indicating a correspondence relationship between the candidate for the diagnostic work and the candidate for the treatment work is displayed. The work contents stored in the diagnostic work content field 124d of the diagnostic work data table 124a and the treatment work content field 123d of the treatment work data table 123a are displayed. Furthermore, an arrow line is connected between the candidate for the diagnostic work and the candidate for the treatment work in which the conditional probability is stored in the conditional probability column 125e of the conditional probability data table 125a. This indicates that the result of the diagnostic work particularly affects the recovery probability of the treatment work connected by the arrow line.

The optimum work calculation processing unit 113 ends the optimum work calculation process when presenting a work with the minimum expected work time value.
Thereby, all the processes in the flowchart of the optimum work calculation process in FIG. 7 and the process of S201 in the flowchart of the work procedure in FIG. 1 are completed. Hereinafter, the processing will be described by returning to the flowchart of the work procedure in FIG.
<Description of Operation of Work Support Main Processing Unit 111>
In step S102, the worker confirms and executes the work displayed in D21 of FIG.
In step S103, the operator further selects the result of the performed work with the selection button D24 and inputs it by pressing the input button D25.

  In step S104, when the operator performs a treatment work to eliminate the abnormality and recover from the abnormality, the process proceeds to the end of the work (S105). If the abnormality has not been resolved, the process returns to the execution of the next operation (S102) again.

  In step S202, when the operator performs a treatment work and inputs information indicating that the abnormality has been resolved, the work support main processing unit 111 proceeds to a data update process (S204). If the abnormality has not been eliminated, that is, if the result of the diagnostic work has been input, or if information indicating that the abnormality has not been eliminated as the result of the treatment work has been entered, the probability update process (S203) The process returns to the optimal work calculation process (S201).

  In step S203, the work support main processing unit 111 performs the following process according to the input results of the diagnosis work and the treatment work, and updates the recovery probability.

(1) (Treatment work) When information indicating that the abnormality has not been resolved as a result of the treatment work is input, the work support main processing unit 111 reads the recovery probability column 123f of the treatment work data table 123a. The information stored in the recovery probability column 123f is updated by normalizing the recovery probabilities of other treatment tasks so that the recovery probability of the executed treatment work is 0 and the sum of the recovery probability field 123f is 1. . (Here, setting the recovery probability of the executed treatment work to 0 is only during the restoration work of the maintenance target device 180 this time, and when the work is completed, in the data update process S204, the relevant treatment work is performed. Will be restored with a new recovery probability.)
(2) (Diagnosis work) When the result of the diagnosis work is input, the work support main processing unit 111 stores it in the recovery probability column 123f of the treatment work data table 123a by the Bayesian update formula of (Equation 3). The update value p (A _s | D _d = R _r ) of the information thus obtained is calculated, and the value is updated.

Here, A _s is a natural number taking s = 1, 2,... S, and indicates a candidate for the treatment work on the sth row of the treatment work data table 123a. S indicates the number of rows.
D _d : Indicates the diagnostic work performed by the worker.
R _r : indicates the diagnosis result input by the operator.
p (A _s ): Indicates a value before update in the s-th row of the recovery probability column 123f.
p (A _s | D _d = R _r ): An update value of the recovery probability when the diagnosis result is input. That is, the update value of the sth row of the recovery probability column 123f of the treatment work data table 123a is shown.
p (D _d = R _r | A _s ): In the table in which the ID2 column 125c of the conditional probability data table 125a is D _d , the diagnostic result column 125d is R _r in the conditional probability column 125e, and the treatment work is A _s. Information stored in the row and column. In the case where there is no treatment operations corresponding to A _s in the rows conditional probability column 125e, information stored in a column of "Others" it is utilized.
A _t is a natural number that takes t = 1, 2,... S, and indicates a treatment work candidate in the t-th row of the treatment work data table 123a in the same manner as s. S indicates the number of rows. The sum symbol indicates that all the treatment work candidates stored in the treatment work data table 123a are summed.
p (A _t ): Similarly to p (A _s ), the value before update of the t-th row in the recovery probability column 123f is shown.
p (D _d = R _r | A _t ): Similarly to p (D _d = R _r | A _s ), in the table in which the ID2 column 125c of the conditional probability data table 125a becomes D _d , the conditional probability column 125e diagnostic result column is R _r of _a information treatment work is stored in rows and columns to be a _t. In the case where there is no treatment operations corresponding to A _t is a column of conditional probability column 125e, information stored in a column of "Others" is utilized.

After updating the probability, the work support main processing unit 111 returns to the optimal work calculation process (S201) and activates the optimal work calculation processing unit 113 again.
FIG. 16 shows an example of the output screen of the optimal work calculation result after the update. When the work A3 is displayed in D31, the operator selects “abnormality has been resolved” in D32, and presses the input button D33, the work support main processing unit 111 performs the data update process (S204).

In step S204, the work support main processing unit 111 updates the information stored in the treatment work data storage area 123 based on the work result. The information stored in the conditional probability data storage area 125 is updated if necessary, but in principle is not updated.
Specifically, the work support main processing unit 111 adds 1 to the line of the treatment work in which the abnormality is eliminated in the recovery number column 123e of the treatment work data table 123a. Further, by dividing the value stored in the recovery count column 123e by the sum of the values stored in the recovery count column 123e, the recovery probability value is calculated and stored in the corresponding row of the recovery probability column 123f. Thus, the information stored in the recovery probability column 123f is updated.

Further, the work support main processing unit 111 sets the row and column values corresponding to the treatment work that has solved the abnormality and the information of the executed diagnosis work and the diagnosis result in the conditional probability column 125e of the conditional probability data table 125a. Update the information stored in the conditional probability column 125e by adding a predetermined numerical value (such as 0.01) and normalizing the sum of the values of each row in the added column to be 1. Also good.
By updating the probability information in this way, the expected value calculation can be performed with higher accuracy in the optimum work calculation process when an abnormality occurs next time.

  Note that not only the probability information but also the work time taken for the treatment work and the diagnosis work is recorded, and the information stored in the treatment work time column 123g of the treatment work data table 123a and the diagnosis of the diagnosis work data table 124a are recorded. The information stored in the work time column 124e may be updated.

FIG. 17 illustrates a configuration example in which the work support apparatus 100 according to the first embodiment illustrated in FIG. 2 is implemented as a cloud system. As illustrated in FIG. 17, the work support system 200 includes an arithmetic device 202, an input unit 260, an output unit 270, and a communication interface unit (communication IF unit) 250, which are mutually connected via a bus 240. It is connected to the.
In addition, the work support system 200 is connected to a maintenance target device 180 that is a diagnosis target via a network 170.

  The computing device 202 includes a computing unit 210 equivalent to the computing unit 110 included in the work support device 100 according to the first embodiment, a storage unit 220 equivalent to the storage unit 120 included in the work support device 100 according to the first embodiment, and the bus 240. An interface unit (IF unit) 230 for transmitting and receiving information via the network is provided.

The input unit 260, the output unit 270, and the communication interface unit (communication IF unit) 250 constitute a client terminal installed at a site where an operator performs maintenance work.
Note that the computing device 202 may be independent of the work support device and connected to the input unit 260 and the output unit 270 via the network 170.

  Although the present invention has been specifically described above based on the embodiments, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the invention.

DESCRIPTION OF SYMBOLS 100 Work support apparatus 110 Operation part 111 Work support main process part 112 Database registration process part 113 Optimal work calculation process part 120 Storage part 121 Work support program storage area 122 Abnormality list data storage area 122a Abnormality list data table 122b ID0 column 122c Abnormal name Field 122d Alarm code field 123 Treatment work data storage area 123a Treatment work data table 123b ID0 field 123c ID1 field 123d Treatment work content field 123e Recovery frequency field 123f Recovery probability field 123g Treatment work time field 124 Diagnostic work data storage area 124a Diagnostic work data Table 124b ID0 column 124c ID2 column 124d Diagnosis work content column 124e Diagnosis work time column 124f Diagnosis result column 125 Conditional probability data storage area 125a Conditional probability data Table 125b ID0 column 125c ID2 column 125d Diagnosis result column 125e Conditional probability column 126 Work sequence data storage area 126a Work sequence data table 126b ID2 column 126c ID3 column 126d Work sequence column 126e Expected value column 126f for each diagnosis result Expected value minimum column 127 Expected value calculation result data storage area 127a Expected value calculation result data table 127b Start work column 127c Expected value column 127 for optimum work order Expected value minimum field 128 Treatment work order data storage area 128a Treatment work order data table 128b Work order field 128c Expected value field 128d Expected value minimum field 130 Input unit 140 Output unit 150 Media reading unit 160 Communication unit 170 Network 180 Maintenance target device 190 Portable terminal 200 Work support system 02 computing device 210 computing unit 220 storage unit 230 interface (IF part)
240 Bus 250 Communication interface section (communication IF section)
260 Input unit 270 Output unit

Claims (12)

A work support device that calculates a candidate for a treatment work necessary for recovering a maintenance target device from an abnormality, or a candidate for a diagnosis work for identifying a treatment work,
A list table of abnormality IDs assumed for the maintenance target device;
A list of treatment work candidates for recovery for each abnormality ID, a recovery table for each treatment work, a recovery probability, a work time data table,
A list of diagnosis work candidates for each abnormality ID, a work time for each diagnosis work, a data table of diagnosis result candidates,
A data table of conditional probabilities indicating the probability of obtaining the diagnosis result of the diagnostic work when the abnormality ID and the candidate of the treatment work for each diagnostic work can be recovered;
A storage unit for storing
An operation sequence for accepting an input of an abnormality ID of the maintenance target device, reading out the candidate for the treatment work and the candidate for the diagnosis work corresponding to the abnormality ID, and continuing the treatment work after each diagnosis work is first executed. Calculate the work order with the minimum expected value of the work time from the combination to the recovery and the work order with the minimum expected value of the work time until the recovery from the combination of the work order to continue the treatment work, An optimum work calculation processing unit that outputs the first diagnostic work in the work order with the smallest expected work time or the treatment work as the optimum work;
A work support apparatus comprising: Instead of the work time of each work stored in the storage unit, work cost, or work man-hours are stored,
Instead of calculating the expected value of work time until recovery in each work order, the optimal work calculation processing unit replaces the expected value of work cost until recovery in each work order or work until recovery in each work order. The work support apparatus according to claim 1, wherein an expected value of man-hours is calculated.   The optimum work calculation processing unit identifies a work order that minimizes an expected work time for each diagnosis result, thereby combining the work orders in which the treatment work is continued after each diagnosis work is first executed. The work support apparatus according to claim 1, wherein a work order with a minimum expected value of work time from the start to the recovery is calculated. It further includes a work support main processing unit,
The work support main processing unit accepts that the worker inputs the result of the execution of the optimum work calculated by the optimum work calculation processing unit and presented to the worker. If the abnormality has not been resolved, the optimal work calculation processing unit is updated to update the recovery probability of the candidate treatment work corresponding to the abnormality ID stored in the storage unit and calculate the next optimal work. The work support apparatus according to claim 1, which is activated again. It further includes a work support main processing unit,
The work support main processing unit accepts that the worker inputs the result of the execution of the optimum work calculated by the optimum work calculation processing unit and presented to the worker. When the abnormality is resolved, in the data table of the treatment work corresponding to the abnormality ID stored in the storage unit, 1 is added to the number of times of restoration of the treatment work that solved the abnormality, and each treatment work on the data table The work support apparatus according to claim 1, wherein an update process is executed for the recovery probability based on the corrected data of the number of times of recovery, and the work support process is terminated. The process of calculating the work order in which the expected value of the work time until recovery from the combination of work orders that continues the treatment work after first executing each diagnosis work executed by the optimum work calculation processing unit is as follows:
Select one of the diagnostic work candidates one by one, create a combination of work orders that can be taken for the treatment work corresponding to the abnormality ID that follows, as the work to perform the selected diagnostic work first Calculate the expected value of the work time until recovery for each diagnosis result for each work order, identify the minimum work order of the expected value for each diagnosis result, and sum the minimum expected value for each diagnosis result The work support apparatus according to claim 1, wherein an expected value of work time when the work is performed in an optimum work order from the selected diagnosis work is calculated. The process of calculating the work order with the minimum expected value of the work time until recovery from among the combination of work orders that continue the treatment work executed by the optimum work calculation processing unit,
The combination of possible work orders is created for the treatment work candidates corresponding to the abnormality ID, and the work order that minimizes the expected value of work time is specified. The work support apparatus described.   When selecting candidates for the treatment work corresponding to the abnormality ID, some treatment work having a recovery probability equal to or higher than a predetermined threshold is selected instead of selecting all the treatment works stored in the data table of the storage unit. The work support apparatus according to claim 1, wherein the work support apparatus is limited to: The work support apparatus according to claim 1,
The work support apparatus is characterized in that the treatment work is associated with a plurality of diagnosis work with a conditional probability. A work support method for calculating a candidate for a treatment work necessary for recovering a maintenance target device from an abnormality using information processing by a computer, or a candidate for a diagnosis work for identifying a treatment work,
A list table of abnormality IDs assumed for the maintenance target device;
A list of treatment work candidates for recovery for each abnormality ID, a recovery table for each treatment work, a recovery probability, a work time data table,
A list of diagnosis work candidates for each abnormality ID, a work time for each diagnosis work, a data table of diagnosis result candidates,
A data table of conditional probabilities representing the probability of obtaining the diagnostic result of the diagnostic work when the abnormality ID and the diagnostic work candidate can be restored for each diagnostic work, and stored in advance in the storage unit,
Receiving an input of an abnormality ID of the maintenance target device, and reading out the candidate for the treatment work and the candidate for the diagnosis work corresponding to the abnormality ID;
From the combination of work sequences in which the treatment work is continued after each diagnosis work is first executed to the recovery from the work order in which the expected value of the work time until recovery is the minimum and the work order in which the treatment work is continued A work support method characterized by calculating a work order with the smallest expected value of work time and presenting the first diagnostic work or treatment work in the work order with the smallest expected value of work time as the optimum work . After presenting the optimum work, the worker accepts the operator to input the result of the execution of the optimum work,
If the abnormality of the maintenance target device has not been resolved, in order to calculate the next optimum operation by updating the recovery probability of the candidate treatment operation corresponding to the abnormality ID stored in the storage unit Repeat the optimal work calculation process,
If the abnormality of the maintenance target device is resolved, in the data table of the treatment work corresponding to the abnormality ID stored in the storage unit, 1 is added to the number of times of restoration of the treatment work that solved the abnormality, The work support method according to claim 10, wherein the recovery support probability of each treatment work on the data table is updated based on the corrected data of the number of times of recovery, and the work support process is terminated. Instead of the work time of each work stored in the storage unit, work cost, or work man-hours are stored,
In the process of calculating the optimum work, instead of calculating the expected value of work time until recovery in each work order, the expected value of work cost until recovery in each work order, or until recovery in each work order The work support method according to claim 10, wherein an expected value of work man-hours is calculated.

Images