JP2013029881A - Maintenance support system, maintenance support apparatus and maintenance support program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a maintenance support system in which optimal work procedures are presented by calculating an expected work time or expected costs on the basis of a maintenance support Tree if any trouble occurs in a maintenance target device.SOLUTION: A maintenance support system comprises a maintenance support Tree information management section, an optimal work calculation section and a diagnostic work execution section. The maintenance support Tree information management section manages maintenance support Tree information containing candidates of treatment work for recovery with respect to a troublesome phenomenon and information on diagnostic work for specifying that treatment. The optimal work calculation section uses the number of cases in the past, experience values of maintenance workers and information of costs required for the diagnostic work or each work in the maintenance support Tree to calculate such a work starting point as to minimize an expected value of the work costs or the work time. In order to repeat the calculation of the optimal work until completing the recovery of a maintenance target device, the diagnostic work execution section updates the maintenance support Tree information while being reflected with a result of the executed work.

Description

  The present invention relates to a maintenance support system, a maintenance support apparatus, and a maintenance support program for supporting maintenance work of a manufacturing apparatus, an inspection apparatus, and the like.

  In order to maintain the quality of manufacturing equipment and inspection equipment over a long period of time, it is necessary to appropriately perform maintenance work such as replacement of parts periodically or when a failure occurs. Also, for service forms that guarantee a certain level of availability and quality in maintenance contracts, maintenance work such as fault diagnosis and maintenance support should be performed accurately and efficiently in the event of a failure, and necessary parts replacement, etc. Improving service levels by reducing downtime due to reducing costs will lead to an improvement in order competitiveness, and at the same time, measures such as reducing maintenance parts costs and work costs will directly improve profitability.

  For example, in failure recovery work for problems caused by wear and deterioration of medical devices, service personnel and operators are required to restore maintenance target devices from failure phenomena in order to identify necessary measures such as parts replacement and adjustment. A general method is to follow a predetermined work procedure described in a maintenance support tree in which treatment work candidates and diagnosis work for specifying treatment work candidates are configured in a tree shape. However, with this method, it is possible to comprehensively investigate necessary treatment candidates, but there is a problem of reducing the time and cost required to identify the treatment.

  As a background art in the technical field to solve this problem, there is Patent Document 1. This gazette collects condition information such as operation status information and usage environment information related to a maintenance target device by using sensor data acquired from a sensor installed in the maintenance target device. Based on this, a system is described in which the maintenance content of the maintenance target device is determined and provided to a maintenance worker. In this system, it is described that by utilizing sensor data, the operation status of each part of the maintenance target device can be grasped and the contents of maintenance corresponding thereto can be presented.

  Patent Document 2 calculates the amount of change in the hit probability, which is the probability that the maintenance target device will be restored by the candidate treatment work when a person or machine performs a failure diagnosis work according to the maintenance support tree. It is described that it is possible to issue an instruction to perform work in order from work with the highest change amount, that is, work with the highest work efficiency.

JP 2005-122707 A JP 2007-193456 A

  The following problems remain in the prior art. The technique of Patent Document 1 is assumed to be used for the determination of limited inspection items to be performed in periodic inspections, and for all necessary treatment work that is assumed when a failure of the maintenance target device is recovered. On the other hand, it is not realistic to associate the value of sensor data as a condition, and in many cases, failure diagnosis must be performed according to the maintenance support tree. In addition, the technique of Patent Document 2 is based on the premise that the results of each diagnosis work are accurate, and does not consider the possibility that the results of the diagnosis work are incorrect, so the assumed diagnosis route is limited. .

  Therefore, an object of the present invention is to provide maintenance support Tree in a large-scale failure diagnosis in which it is difficult to specify a necessary recovery procedure on the condition of only the sensor data value when a failure occurs in the maintenance target device. A maintenance support program that calculates expected work time or expected cost in consideration of the ambiguity of diagnosis work results and presents optimum work procedures to service personnel and operators, and a system that executes this method are provided. That is.

In order to solve the above problems, for example, the configuration described in the claims is adopted. The present application includes a plurality of means for solving the above problems.
A maintenance support system for supporting a maintenance work by presenting a candidate for a treatment work necessary for a maintenance work of a maintenance target device or a candidate for a diagnostic work for specifying a necessary treatment work. Using the maintenance support tree information management unit that manages the maintenance support tree information in which the candidates and the candidates for the diagnosis work are configured in a tree shape, and the cost information of each work that constitutes the maintenance support tree information, the work cost or the work time An optimal work calculation unit that calculates the optimal work that minimizes the display, and a display unit that presents the calculated optimal work, and the maintenance support tree information management unit reflects the result of the executed treatment work Then, the maintenance support tree information is updated, and the optimum work calculation unit calculates the optimum work based on the updated maintenance support tree information. And butterflies.

In addition, the fault diagnosis system of the present application minimizes the expected value of the work cost or work time under the condition that the maintenance worker performs the calculated optimum work at the beginning of the maintenance work. In the case where the assumed work simulation unit further calculates one or more assumed work to be performed after the optimum work, and the assumed work simulation unit is the lowest-level treatment work. Are excluded from the candidates for the assumed work calculation, or if they are not the lowest-level treatment work, case points N _Y , N of the next lower hierarchy compared to _N, updating excluded maintenance support Tree information work or treatment operations basic building blocks of a smaller value side of the case point from candidates of the assumed work calculation And performing a physical.

  According to the present invention, in a large-scale failure diagnosis when a failure occurs in a maintenance target apparatus, an expected work time or an expected cost is calculated in consideration of the ambiguity of the diagnosis work result based on the maintenance support tree. This enables maintenance workers to be presented with the optimal maintenance support work procedure according to the situation, and by reducing the downtime of the maintenance target device by speeding up the work, the service level can be improved and work costs and inventory costs can be reduced. There is a reduction effect.

It is a figure explaining the maintenance service model of one Embodiment of this invention. It is a system function block diagram of one Embodiment of this invention. It is a figure explaining the hardware constitutions of the information terminal of a maintenance assistance system. It is a figure explaining the processing step of the system of one embodiment of the present invention. It is a flowchart of defect information reception processing. It is a flowchart of optimal work calculation processing (1). It is a flowchart of optimal work calculation processing (2). It is a flowchart of an assumed work simulation process. It is a flowchart of a diagnostic work execution process. It is a figure explaining the data structure of the reception information storage part (A) 41. It is a figure explaining the data structure of the reception information storage part (B) 42. It is a figure explaining the data structure of the maintenance work information storage part 43. FIG. It is a figure explaining the data structure of the maintenance assistance tree master information storage part 44. FIG. It is a figure explaining the data structure of the maintenance assistance tree simulation information storage part 45 and the maintenance assistance tree execution information storage part 46. FIG. It is a figure explaining the data structure of the maintenance components information storage part. It is a figure explaining the data structure of the maintenance work record information storage part. It is a figure explaining the data structure of the sensor information storage part. 4 is a diagram for explaining a data structure of an alarm data storage unit 50. FIG. It is a figure explaining the example which divides | segments maintenance assistance Tree information according to the hierarchy of a basic structural block. It is a figure which shows the example which displayed the maintenance support result on the user interface. It is a figure explaining the example of a component of the basic composition block of maintenance assistance Tree.

Hereinafter, an embodiment of a maintenance support system according to the present invention will be described.
1. Maintenance Service Model of One Embodiment of the Present Invention A maintenance service model assumed as an application target of the maintenance support system of the present embodiment will be described with reference to FIG. The main object of the present invention is medical equipment such as an MRI apparatus, an X-ray CT apparatus, an X-ray diagnostic apparatus, and an ultrasonic diagnostic apparatus. However, the object is not limited to this example, and the present invention can be applied to manufacturing apparatuses such as power generation facilities, plants, construction machines, aircraft, automobiles, elevators, and inspection apparatuses such as an optical microscope.

  At the support center ε, when the maintenance target device is determined to be in an abnormal state periodically or based on the trouble report 27 or the sensor data 28 reported from the users of the maintenance target device installation sites β1, β2,. Whether the maintenance work is necessary or not is determined using the alarm information 26 automatically issued as a trigger. When maintenance work is required, the parts delivery instruction 22 for the maintenance parts 52 is issued from the warehouse γ to the device installation sites β1, β2,..., Βm to be maintained, and the service bases α1, α2,. The maintenance worker dispatch instruction 24 of the maintenance worker 51 is performed. Here, the service bases α1, α2,..., Αm are arranged for each region, and each is responsible for maintenance work of a plurality of maintenance target device installation sites β1, β2,.

The entry / exit record 21 and the maintenance work record 23 of the maintenance parts 52 are registered in the maintenance support system in the warehouse γ and the service bases α1, α2,..., Αm, respectively, and the support center ε is utilized by utilizing the stored maintenance case information. Generate maintenance support rules. The failure diagnosis support information 25 based on the maintenance support rule is used to support maintenance work, so the support center ε contacts the maintenance worker 51, and the maintenance worker 51 replaces the failure diagnosis with reference to the failure diagnosis result 25. The maintenance parts 22 to be selected are selected.
2. System Configuration of One Embodiment of the Present Invention A configuration of a maintenance support system of the present embodiment will be described with reference to FIG.
This maintenance support system requires that the maintenance target device be restored when a failure due to wear or deterioration occurs in the maintenance target device or when a maintenance worker finds a sign of failure during regular inspection. In case of large-scale failure diagnosis, based on the maintenance support tree, calculate the expected value of work time and work cost considering the ambiguity of the diagnosis work result, and support the optimal work procedure to minimize them This system is presented to the center operator and local maintenance workers.

  The maintenance support system manages the maintenance support tree and calculates the optimum diagnosis work procedure based on the maintenance support tree when a failure occurs, and the price and delivery of the maintenance parts used for the maintenance work. Maintenance part information management module 12 that manages costs and inventory status, maintenance work record information management module 13 that manages work record information of maintenance work that has been performed in the past, and the results of fault diagnosis when a failure occurs The display terminal module 14 to be output to the maintenance work support terminal to be presented to the worker and the data acquired from each sensor installed in the maintenance target device are managed, and when the sensor data indicates an abnormal value, the support center The sensor data management module 15 is configured to transmit alarm information to the sensor data management module 15.

  As shown in FIG. 1, each module includes an optimum diagnosis work calculation module 11 and a display terminal module 14 at a support center ε, a maintenance part information management module 12 at a warehouse γ, and service bases α1, α2,. The maintenance work record information management module 13 includes a display terminal module 14 in an information terminal carried by the maintenance worker 21 and a sensor data management module 15 in maintenance target device installation sites β1, β2,. Each component module 11-15 is connected to the network 71, and each information terminal 11-15 can transmit / receive various data etc. via the network 71 mutually.

  As shown in FIG. 3, each of the component modules 11 to 15 is a computer, and includes an input device 61 such as a keyboard and a mouse, an output device 62 such as a display, an auxiliary storage device 63, and a maintenance support program. And an arithmetic device 60 that executes the program. The arithmetic device 60 includes a central processing unit (hereinafter referred to as CPU) 64, a main storage device 65, and an interface 66. The arithmetic device 60 is connected to an input device 61, an output device 62, and an auxiliary storage device 63 through an interface 66.

  In the present embodiment, execution results of various programs such as a maintenance support program are stored in a storage area secured in the main storage device 65. Various programs are stored in advance in the auxiliary storage device 63, then read into the main storage device 65, and executed by the CPU 44. Various functions described later are realized by the execution of various programs by the CPU 64.

  In the present embodiment, a case where each information terminal constituting the maintenance support system is realized by a general-purpose information processing apparatus and software will be described as an example. For example, hardware including hard-wired logic or such hardware And a general-purpose information processing apparatus programmed in advance.

  In the present embodiment, the maintenance support system is described as an integrated system, but the present invention is not limited to this. The present invention may be configured to be incorporated in other information processing systems and function as a part thereof. Further, a part of each information terminal function may be rearranged, subdivided, or combined.

Next, the functional configuration of each information terminal 11 to 15 constituting the maintenance support system and the data held by each information terminal 11 to 15 will be described with reference to FIGS. 2 and 10 to 18.
3. Functional Configuration and Retained Data Contents of One Embodiment of the Present Invention Each component module 11 to 15 of the maintenance support system of the present embodiment, as shown in FIG. 40 functional units and 41 to 50 storage units for storing various data.

The arithmetic unit is
A reception information management unit 31 for managing the relationship between alarm information at the time of a malfunction or a report from a user and a maintenance support tree corresponding to the alarm information;
Maintenance for managing maintenance support tree information configured in a tree form by adding connection information before and after the candidate for the work required for recovery for each malfunction phenomenon and the information of the diagnostic work for identifying it. A support tree information management unit 32;
An optimum work calculation unit 33 for calculating a work start point at which an expected value of work cost or work time is minimized during the maintenance support tree;
An assumed work simulation unit 34 for calculating a plurality of assumed work to be performed after the optimum work in order to minimize the expected value of work cost or work time when the result of the optimum work is assumed;
A maintenance part information management unit 35 for managing cost information and inventory status of maintenance parts used for maintenance work;
A maintenance work record information management unit 36 for managing work record information of maintenance work performed in the past;
A sensor data management unit 37 for managing data acquired from each sensor installed in the apparatus to be maintained;
An alarm data management unit 38 for transmitting alarm information to the support center when the sensor data indicates an abnormal value;
A failure diagnosis result output unit 39 for outputting information for supporting a maintenance worker's failure diagnosis work to the maintenance work support terminal of the service center;
A diagnostic work execution unit 40 that updates the maintenance support tree information by reflecting the result of the executed diagnostic work.

  As described above, each of the functional units 31 to 40 realized by the arithmetic device functions when the CPU 64 executes various programs. Details of operations of these functional units will be described later in the description of the processing flow.

The storage unit
An acceptance information storage unit (A) 41 that stores alarm information when a malfunction occurs in the maintenance target device and corresponding information of the maintenance support tree corresponding thereto;
A reception information storage unit (B) 42 in which correspondence information of a keyword included in a report from a user when a malfunction occurs in a maintenance target device and a maintenance support tree corresponding to the keyword is stored;
Maintenance work information storage unit 43, which is a work area for evaluating work with the highest work efficiency by disassembling maintenance support tree information into basic structural blocks by hierarchy;
A maintenance support tree master information storage unit 44 in which a maintenance work tree including information on a candidate treatment work necessary for recovery for each malfunction phenomenon and diagnostic work for specifying the candidate is stored;
A maintenance support tree simulation information storage unit 45 for storing maintenance support trees when calculating a plurality of assumed tasks to be performed after the optimum operation;
A maintenance support tree execution information storage unit 46 storing a maintenance support tree reflecting the obtained diagnosis result when the maintenance worker advances the fault diagnosis work;
A maintenance part information storage unit 47 for storing cost information and inventory status of maintenance parts used for maintenance work;
Maintenance work record information storage unit 48 for storing information on the contents of maintenance work performed in the past;
A sensor data storage unit 49 for storing data of sensors installed to monitor the operation status of the maintenance target device;
And an alarm data storage unit 50 for storing alarm data transmitted when the sensor data indicates an abnormal value.

  As shown in FIG. 10, the reception information storage unit (A) 41 is created in advance corresponding to a failure phenomenon and an alarm ID field 41a for storing an alarm ID for identifying at least a failure alarm issued from the maintenance target device. It has a maintenance support tree ID field 41b for storing a maintenance support tree ID for specifying the maintenance support tree, and a failure phenomenon explanation field 41c for storing display information for explaining the failure phenomenon.

  As shown in FIG. 11, the reception information storage unit (B) 42 corresponds to a failure keyword and a notification keyword field 42a for storing a keyword included in a notification from a user who conveys at least a failure phenomenon of the maintenance target device. It has a maintenance support Tree ID field 42b for storing a maintenance support Tree ID for specifying a maintenance support Tree created in advance, and a failure phenomenon description field 42c for storing display information for explaining a failure phenomenon.

As shown in FIG. 12, the maintenance work information storage unit 43 has at least
A basic configuration block ID field 43a for storing ID information for identifying each basic configuration block (m, j) in the data record of each basic configuration block from the lowest layer to the upper layer in order from the maintenance support Tree master information;
A work ID for storing ID information for identifying a diagnostic work, a treatment work, or a basic constituent block one level below, constituting the basic constituent block (m, j), a lower block ID field 43b,
A cost field 43c for storing cost information of diagnostic work and treatment work constituting the basic constituent block (m, j) or representative work of the basic constituent block of the next lower hierarchy;
An experience value field 43d for storing an experience value in which a value of the case occurrence frequency of the corresponding work that the person in charge of performing the failure diagnosis has obtained through experience is set by subjective judgment;
A number-of-examples field 43e for storing the number of cases in which the maintenance support system has treated the corresponding work that has been accumulated as a result so far;
A case point field 43f for storing a value N obtained by calculating a case point N = e + kn (k is a case coefficient and an arbitrary constant) obtained by combining the experience value e and the number n of cases;
The determination certainty factor is a probability that the result determined as Yes or No in the diagnosis work is a correct probability, and is set statistically from the past work results, or stored by the maintenance worker arbitrarily set from the experience. Certainty field 43g,
When it is assumed that the diagnosis process is started from the above-described work ID, diagnosis work stored in the lower block ID field 43b, treatment work, or representative work of the basic configuration block one level below, the calculation is performed by the calculation method described later. An expected cost field 43h for storing the expected cost;
An optimal work field 43i for storing the corresponding work ID is referred to as an optimal work, and the work having the lowest expected cost among the expected costs calculated on the assumption that each of the above works is the starting point of the diagnostic process;
A predetermined number field 43i for storing a predetermined number set by the user for the number of assumed tasks to be described later.

As shown in FIG. 13, the maintenance support tree master information storage unit 44 has at least
A work ID field 44a for storing a work ID for identifying each diagnostic work and each treatment work; and a data record representing the contents of each diagnostic work and each treatment work constituting the maintenance support tree;
A work attribute field 44b for identifying a diagnosis work or a treatment work;
A work name field 44c for storing the name of the work;
A work content / determination method field 44d for storing text data explaining the work content or the judgment method;
A part number field 44e for storing a part number of a part used in the corresponding work;
A work mode field 44f for storing information for identifying whether the subject performing the work is a maintenance worker or a sensor;
An experience value field 44g for storing data in which an experienced worker registers the experience value for the corresponding work in advance, or a maintenance worker registers the experience value for the corresponding work prior to the diagnosis process,
A work cost field 44h for storing a work cost that is statistically calculated from past performance data for performing the work;
When the corresponding work is a diagnostic work, a determination certainty field 44i that stores the above-described determination certainty that the result determined as Yes or No is a correct probability;
If the corresponding work is a diagnostic work, the work ID or the end work ID of the treatment work to be performed after determining Yes or No is stored, and if the corresponding work is the treatment work, the work ID is subsequently executed. And a Next work field 44j for storing a work ID of a diagnosis work or an end work ID.

The maintenance support tree simulation information storage unit 45 and the maintenance support tree execution information storage unit 46 have substantially the same fields as the maintenance support tree master information storage unit 44, as shown in FIG. It has 46k in addition.
The configuration of these data records includes at least work ID fields 45a and 46a, work attribute fields 45b and 46b, work name fields 45c and 46c, work content / determination method fields 45d and 46d, and part number fields 45e, 46e, work mode fields 45f and 46f, experience values 45g and 46g, work cost fields 45h and 46h, determination certainty fields 45i and 46i, Next work fields 45j and 46j, and the corresponding work are excluded from the calculation target. Calculation target flag fields 45k and 46k for storing determination flags as to whether or not to perform.

  As shown in FIG. 15, the maintenance part information storage unit 47 includes at least a part number field 47a for storing a part number for specifying a part, a part unit price field 47b for storing a part unit price, and a part delivery cost. It has a delivery cost field 47c for storing, and an inventory quantity field 47d for storing the inventory quantity of parts.

  As shown in FIG. 16, the maintenance work record information storage unit 48 records each data record as a record of performance every time a maintenance worker performs a maintenance work. Each data record includes at least a maintenance ID field 48a for storing a maintenance ID for identifying a maintenance work record, a site number field 48b for recording a site for performing the maintenance work, and a work date field 48c for recording a work day. Product number field 48d for recording the product number to be diagnosed, alarm type / keyword field 48e for recording an alarm issued from the product to be diagnosed or a keyword included in a report from the user, and attributes of trigger information for diagnosis processing , A maintenance support tree field 48g for recording the ID of the employed maintenance support tree, and a work ID field 48h for recording the ID of the maintenance work performed.

  As shown in FIG. 17, the sensor data storage unit 49 includes a product number field 49a for storing information for specifying a maintenance target device, and a site number field for storing information for specifying a site where the maintenance target device is installed. 49b, a date / time field 49c for storing date / time information for collecting sensor information, and a sensor number field 49d for storing each sensor data.

As shown in FIG. 18, the alarm information storage unit 50 stores a product number field 50a for storing information for specifying a maintenance target device, and a site number field for storing information for specifying a site where the maintenance target device is installed. 50b, a date / time field 50c for storing the date / time when alarm data is received, and an alarm type field 50d for recording the type of alarm.
4). Flow of the entire process of the maintenance support system As shown in FIG. 4, the process performed by the maintenance support system is as follows: (1) When a problem occurs in the maintenance target device, an alarm sent from the maintenance target device or a report from the user From the contents of the received information, the failure information receiving process S1 for selecting the target maintenance support tree, and (2) calculating the work start point at which the expected value of the work cost or the work time is minimized during the maintenance support tree. Optimal work calculation processing S2, and (3) When assuming the result of optimal work, in order to minimize the expected value of work cost or work time, a plurality of assumed work to be performed after the optimal work is calculated. Assumed work simulation process S3, and (4) assisting maintenance personnel to perform fault diagnosis while using the maintenance support system during maintenance work. The diagnosis execution treatment S4, we are.

Further, the execution result of the maintenance support system is output to the maintenance work support terminal 11 for presentation to the support center operator. The details of the above processing will be described in detail using a flowchart.
5. Defect information reception process S1
The process flow of the defect information acceptance process S1 will be described with reference to the flowchart shown in FIG.

  First, the alarm information received from the maintenance target device is acquired from the alarm data storage unit 50, or the failure notification notified from the user is registered in the maintenance work record information storage unit 48 as trigger information (S101).

  It is determined whether the trigger information is an alarm report or a report from the user (S102).

  If the trigger information is an alarm notification, the reception information storage unit (A) 41 is referred to, a maintenance support tree corresponding to the failure phenomenon is selected and registered in the maintenance work record information storage unit 48 (S103). .

When the trigger information is a report from the user, the reception information storage unit (B) 42 is referred to, a maintenance support tree corresponding to the malfunction phenomenon is selected and registered in the maintenance work record information storage unit 48 (S104). ).
6). Optimal work calculation process S2
In the conventional optimal work calculation process S2, when the diagnosis process is performed using the maintenance support tree, the diagnosis process is started from the diagnosis work of the highest hierarchy. In order to solve a problem that requires time to reach, that is, a cost, it is for finding a place where the diagnosis process is started from the diagnosis work / treatment work that can be regarded as optimum.

  Therefore, in the optimum work calculation process S2, the basic components such as the diagnosis work and the treatment work that constitute the maintenance support tree are combined and attention is paid to the basic structure block as shown in FIG. The basic configuration block is composed of three tasks: one diagnosis work (Yes or No two-branch determination process) and two treatment works in the lowest layer of the maintenance support tree and the hierarchy one level above it. The three tasks are represented by the symbol D for the upper level diagnosis task, the task Y for the lower layer AY, and the task No for the lower layer AN. Then, one basic component block is regarded as one representative treatment operation, and a higher-level basic component block that is hung from the diagnosis operation of the next higher hierarchy is recursively configured. Then, a basic configuration block is configured that is hierarchized up to the basic configuration block including the highest level diagnosis work. However, it is assumed that the treatment work hanging on the diagnosis work above the basic composition block at the lowest layer constitutes the basic construction block by one treatment work.

  The flow of the optimal work calculation process S2 will be described with reference to the flowcharts shown in FIGS.

First, for the maintenance support tree selected in the defect information reception process S1, the maintenance support tree master information storage unit 44 is read, and as described above, the maintenance support tree master information block 44 is disassembled into basic configuration blocks for each hierarchy, and each maintenance work information storage unit 43 stores A data record area 430, a basic configuration block ID 43a, a work ID, and a lower block ID 43b of the basic configuration block are registered. (S201)
Subsequently, 1 is assigned as the initial value to the block hierarchy (m) representing the hierarchy of the basic component block, and 1 is assigned as the initial value of the block number (j) for each hierarchy. Here, in the block hierarchy (m), the lowest layer of the maintenance support tree is set to 1, and the value increases as the hierarchy becomes higher. The block number (j) for each hierarchy is a serial number assigned to the basic configuration block for each hierarchy (S202).

Next, with respect to the selected basic configuration block (m, j), the maintenance support tree master information storage unit 44 uses the operation ID and lower block ID 43b of the component registered in the maintenance operation information storage unit 43 as search keys. The part number, the work cost, the experience value, and the determination reliability are acquired, and the experience value 44g is registered in the experience value (e _Y , e _N ) 43d of the maintenance work information storage unit 43, and the determination reliability ( _py , p _N ) The determination certainty factor 44i is registered in 43g. The determination certainty is the probability that the result determined as Yes or No in the diagnosis work is correct as described above, and is a value set statistically from the past work results or arbitrarily set by the maintenance worker based on experience It is. Then, the part unit price and the delivery cost of the part number acquired from the maintenance support tree master information storage unit 44 are acquired from the maintenance part information storage unit 47, and the work cost (C _D , C _AY , C _{AN of the} maintenance work information storage unit 43 is acquired. ) 43c is registered with the value of work cost 44h + part unit price 47b + delivery cost 47c. Furthermore, the result totaled for each work ID is acquired from the maintenance work record information storage unit 48 and registered in the number of cases (n _Y , n _N ) 43e in the maintenance work information storage unit 43 (S203).

  When the treatment work AY, AN of the component of the basic configuration block (m, j) registered in the maintenance work information storage unit 43 is a basic configuration block ID of a lower hierarchy, the corresponding basic configuration block of the lower hierarchy Referring to the column of the optimum work 43i, corresponding data is read from the maintenance support tree master information storage unit 44 using the optimum work ID as a search key and registered in the maintenance work information storage unit 43.

Next, the expected cost (EC _AY , EC _AN , EC _D) required when it is assumed that the operations of AY, AN, D, which are the components of the selected basic building block (m, j), are first started. ) And is registered in the expected cost (EC _AY , EC _AN , EC _D ) 43h of the maintenance work information storage unit 43 (S204).

The expected cost is calculated by the following method when any of the operations constituting the basic building block is performed first.
All routes for diagnosing the diagnostic work D and the treatment work AY, AN, which are the components of the basic structural block shown in FIG. 21, are defined as the following eight routes [1] to [8]. EC _AY , EC _AN , and EC _D are calculated by the following calculation formulas, considering the possibility of passing through the diagnostic processing route from [1] to [8]. The case point (N) = experience value (e) + k · number of cases (n) is calculated, and the case coefficient (k) is an arbitrary constant.

  [1] {AY work implementation → Restored and finished. } Expected cost

Here, N _Y is a case point of the treatment work AY, and N _N is a case point of the treatment work AN. That is, EC ₁ = (case OK probability) × (measurement cost).

  [2] {AY work implemented → AN work not restored, AN work implemented → Restored and finished. } Expected cost

That is, EC ₂ = (example NG probability) × (sum of countermeasure costs).

Therefore, the expected cost EC _AY required when the AY work is first started is

  [3] {Execute AN work → Restore and finish. } Expected cost

That is, EC ₃ = (case OK probability) × (countermeasure cost).

  [4] {AN work implementation → No recovery, AY work implementation → Recovery and end. } Expected cost

That is, EC ₄ = (case NG probability) × (sum of countermeasure costs).

Therefore, the expected cost EC _AN required when the AN work is first started is

  [5] {D work implementation → AY work implementation → Recovery and end. } Expected cost

Here, P _Y is a determination certainty factor that represents a probability that the result of determination in the diagnosis work D is Yes. That is, EC ₅ = (determination correct answer probability) × (case OK probability) × (sum of determination cost and countermeasure cost).

  [6] {D work implementation → AY work implementation → No recovery, AN operation implementation → Recovery and end. } Expected cost

That is, EC ₆ = (determination error probability) × (case OK probability) × (sum of determination cost and countermeasure cost).

  [7] {D work implementation → AN work implementation → Restored and finished. } Expected cost

Here, _PN is a determination certainty factor that represents a probability that the result of determining No in the diagnosis work D is correct. That is, EC ₇ = (determination correct answer probability) × (case OK probability) × (sum of determination cost and countermeasure cost).

  [8] {Perform D work → Perform AN work → Do not recover, perform AY work → Recover and end. } Expected cost

That is, EC ₈ = (determination error probability) × (case OK probability) × (sum of determination cost and countermeasure cost).

Therefore, the expected cost EC _D applied to the case, which began in the first D work,

Next, among the tasks AY, AN, and D, the task having the minimum expected cost is determined as the optimum task, and the task ID of the optimum task is registered in the optimum task (s _mj ) 43i of the maintenance task information storage unit 43. (S205).

The expected cost of the optimal operation (s _mj ) of the basic configuration block (m, j) is referred to as a representative value of the basic configuration block (m, j). It is assumed that the two basic values of the two basic configuration blocks and the three basic operations blocks of the upper hierarchy are configured by three operations of one diagnostic operation in the upper hierarchy.

  Next, it is determined whether or not the block hierarchy m is the highest hierarchy by searching the work ID and lower block ID column 43b of the maintenance work information storage unit 43. If it is not the highest hierarchy, the process proceeds to the next step. If it is the highest hierarchy, the process proceeds to S211 (S206).

The basic configuration block (m + 1, p) in the upper hierarchy of the basic configuration block (m, j) is identified by searching the work ID and lower block ID fields of the maintenance work information storage unit 43, and the upper hierarchy basic configuration block Substitute min {EC _D , EC _AY , EC _AN } into the cost column 43c of (m + 1, p), and N _Y + N _N into the corresponding column 43f of the case point of the basic component block (m + 1, p) in the upper hierarchy. (S207).

Thereafter, it is determined whether the value of the block number j for each hierarchy is the last number of the corresponding hierarchy (S208),
If the value of the hierarchical block number j is not the last number of the corresponding hierarchy, j + 1 is substituted for j, and the processing from S203 to S1206 is repeated (S209).

When the value of block number j by hierarchy is the last number of the corresponding hierarchy,
Substituting m + 1 for the block hierarchy m and 1 for the hierarchical block number (j), the processes from S204 to S207 are repeated (S210).

When the calculation has been completed up to the basic configuration block of the highest hierarchy, the optimum operation s _mj of the basic configuration block of the highest hierarchy is referred to (S211), and it is determined whether or not s _mj is the diagnosis operation D. (S212).

When s _mj is the diagnosis work D, s _mj is set as the optimum work of the maintenance support tree (S215), and the optimum work calculation process S2 is terminated.

If s _mj is not diagnostic work D, it is determined whether s _mj is the lowest level treatment work A _1Y or A _1N (213);
If s _mj is not the lowest level treatment operation A _1Y or A _1N , if s _mj is AY, s _{(m−1) j} of the block on the YES side of the next lower layer is referred to, and s _mj If AN is AN, s _{(m-1) j} of the NO side block of the next lower layer is referred to, and the processing from S212 to S213 is repeated (S214).

If s _mj is the lowest-level treatment work A _1Y or A _1N , s _mj is set as the optimum work of the maintenance support tree (S215), and the optimum work calculation process S2 ends.

In the present embodiment, an example is shown in which the work cost is used as an evaluation index for calculating the optimum work, but it is also conceivable to use the work time as the evaluation index. In this case, the work cost field 44h of the maintenance support tree master information storage unit 44 is changed to a work time field, the cost field 43c of the maintenance work information storage unit 43 is changed to a time field, and the expected cost of the maintenance work information storage unit 43 is changed. The field 43h is changed to an expected time field, the work cost fields 45h and 46h of the maintenance support tree simulation information storage unit 45 and the maintenance support tree execution information storage unit 46 are changed to work time fields, and the delivery cost field of the maintenance part information storage unit 47 47c may be changed to a delivery time field, data relating to each time may be stored, and work time may be used as an evaluation index.
7). Assumed work simulation process S3
The process flow of the assumed work simulation process S3 for calculating a plurality of assumed works to be performed after the optimum work will be described with reference to the flowchart shown in FIG.

  First, the predetermined number (A) of the assumed work calculated in the assumed work simulation process S3 is set, registered in the maintenance work information storage unit 43 (S301), and 1 is substituted as the initial value of the calculated assumed work number (a). (S312).

Subsequently, if the optimum work calculated in the processing of S1 and S2 or one or more assumed work has already been calculated, it is determined whether or not the assumed work calculated immediately before is the lowest-level treatment work. (S302),
If it is not the lowest-level treatment operation, the basic configuration block of the next lower level hanging on the optimum operation or the assumed operation calculated immediately before, or the case points (N _Y , N _N) on the YES side and NO side of the treatment operation ) (S303)
In the case of N _Y ≧ N _N, the NO-side basic configuration block or treatment work hanging on the optimal work or the assumed work calculated immediately before is excluded from the candidates for the next assumed work calculation, and when N _Y <N _N Excludes the basic configuration block or treatment work on the YES side that hangs on the optimum work or the assumed work calculated immediately before from the candidate for the next assumed work calculation and registers it in the maintenance support tree simulation information storage unit 45 (S304). The maintenance support tree simulation information storage unit 45 is created by copying the contents of the maintenance support tree master information storage unit 44 or the maintenance support tree execution information storage unit 46 at the start of the assumed work simulation process S3. And The maintenance support tree execution information storage unit 46 is created by copying the contents of the maintenance support tree master information storage unit 44. When a failure diagnosis work is executed, the maintenance support tree information reflects the work result. Is updated.

  Here, when the basic configuration block or the treatment work is excluded from the candidates for the next assumed work calculation and registered in the maintenance support tree simulation information storage unit 45, the maintenance support tree simulation information corresponding to each work constituting the basic configuration block. It is to write information of x in the calculation target flag column 45k of the data record on the storage unit. When the data record of the maintenance support tree simulation information storage unit 45 is created, “◯” is written as the initial value of the calculation target flag column 45k.

  If the determination result in S302 is the lowest level treatment work, the maintenance work Tree is excluded by diagnosing the optimum work or the diagnosis work in the hierarchy immediately above the assumed work calculated immediately before from the candidate for the next assumed work calculation. The information is registered in the simulation information storage unit 45 (S305).

  Next, the optimum work or the assumed work calculated immediately before is excluded from the candidates for the next assumed work calculation and registered in the maintenance support tree simulation information storage unit 45 (S306), and the maintenance support tree simulation information storage is performed as in S201. The unit 45 is read, and only the data record with the calculation target flag 45k being ◯ is decomposed into basic structural blocks by hierarchy, and new maintenance work information is created in the maintenance work information storage unit 43 (S307). Then, processing similar to S202 to S215 of the optimal work calculation processing S2 is executed to calculate the next assumed work (S308).

  It is determined whether the assumed work number (a) calculated by the assumed work simulation process S3 has reached the predetermined number (A) set in S301 (S309), and if the predetermined number (A) has not been reached, a + 1 is substituted for a, and the processing from S302 to S309 is repeated (S310).

  When the predetermined number (A) is reached, the calculation result is displayed on the failure diagnosis result output unit 39 (S311), and the assumed work simulation process S3 is terminated.

FIG. 20 shows an example displayed on the failure diagnosis result output unit 39. The instruction detail field displays the work contents evaluated as the optimum work and the work contents evaluated as the assumed work, and the maintenance support Tree field on the right side shows the positions of the optimum work a and the assumed works b and c. Is displayed.
8). Fault diagnosis execution procedure S4
The failure diagnosis execution process S4 refers to the diagnosis result of the maintenance support system from the display terminal module 14 when the maintenance worker performs the maintenance work, and sequentially reflects the result of the maintenance work on the maintenance support system. This process is executed by the maintenance support system when an interactive operation is assumed in which the maintenance support system updates the diagnosis result. The maintenance worker sees the diagnosis result as shown in FIG. 20, grasps the next work content, answers the inquiry from the maintenance support system for the maintenance work, and inputs the result of the maintenance work to the input device 61. To enter into the maintenance support system.

  The process flow of the failure diagnosis execution process S4 will be described with reference to the flowchart shown in FIG.

First, the maintenance worker looks at the failure diagnosis result displayed on the display terminal module 14 to determine / input whether the performed work is a treatment work, and the maintenance support system accepts the determination input (S401). ),
If it is not a treatment work, if the determination result of the performed diagnostic work is YES, the NO-side basic configuration block or the treatment work hanging from the performed work is deleted from the maintenance support tree execution information storage unit 46, In the case of NO, the YES-side basic configuration block or treatment work hanging from the work performed is deleted from the maintenance support tree execution information storage unit 46 (S402),
If it is a treatment work, it is determined whether the restoration of the maintenance target device is completed as a result of the treatment work (S403),
When the restoration of the maintenance target device is completed, the result of the restoration completion is displayed on the failure diagnosis result output unit 39 (S409), and the failure diagnosis execution process S4 is terminated.

  If the restoration of the maintenance target device is not completed, the diagnostic work of the hierarchy one level higher than the performed work is deleted from the maintenance support tree execution information storage unit 46 (S404).

  Subsequently, the performed work is deleted from the maintenance support tree execution information storage unit 46 (S405).

  Here, deleting the basic configuration block or each work from the maintenance support tree execution information storage unit 46 means that the calculation target flag of the data record on the maintenance support tree execution information storage unit 46 corresponding to each work constituting the basic configuration block It is to write x information in the column 46k. When the data record of the maintenance support tree execution information storage unit 46 is created, “◯” is written as the initial value of the calculation target flag field 46k.

As in S201, the maintenance support tree execution information storage unit 46 is read, and only the data records with the calculation target flag 46k being ○ are decomposed into hierarchical basic configuration blocks, and new maintenance work information is replaced with maintenance work information. Create in the storage unit 43 (S406),
The same processes as S202 to S215 of the optimum work calculation process S2 are executed to recalculate the optimum work (S407).

Next, an assumed work simulation process S3 is executed (S408),
The calculation results of the optimum work calculation process S2 and the assumed work simulation process S3 are displayed on the failure diagnosis result output unit 39 (S409), and the failure diagnosis execution process S4 is terminated.

  The maintenance support tree execution information storage unit 46 is created by copying the contents of the maintenance support tree master information storage unit 44 at the start of the diagnostic work execution process S4, and when the work is executed. The maintenance support tree information is updated to reflect the work result.

11 ... Optimal diagnosis work calculation module, 12 ... Maintenance parts information management module,
13 ... Maintenance work record information management module, 14 ... Display terminal module,
15 ... Sensor data management module,
21, 22, 23, 24, 25, 26, 27, 28 ... data transmitted and received between each base,
31 ... Reception information management unit, 32 ... Maintenance support tree information management unit,
33 ... Optimal work calculation part 34 ... Assumed work simulation part 35 ... Maintenance part information management part 36 ... Maintenance work record information management part 37 ... Sensor data management part 38 ... Alarm data management part 39 ... Failure diagnosis result Output section,
DESCRIPTION OF SYMBOLS 40 ... Diagnostic work execution part, 41 ... Reception information storage part (A), 42 ... Reception information storage part (B), 43 ... Maintenance work information storage part, 44 ... Maintenance support tree master information storage part, 45 ... Maintenance support tree Simulation information storage unit 46 ... Maintenance support tree execution information storage unit 47 ... Maintenance part information storage unit 48 ... Maintenance work record information storage unit 49 ... Sensor data storage unit
50 ... Alarm data storage unit 51 ... Maintenance worker 52 ... Maintenance parts
DESCRIPTION OF SYMBOLS 60 ... Arithmetic unit, 61 ... Input device, 62 ... Output device, 63 ... Auxiliary storage device, 64 ... Central processing unit (CPU), 65 ... Main storage device, 66 interface, 71 ... Network

Claims (10)

A maintenance support system for supporting a maintenance work by presenting a candidate for a work required for a maintenance work of a maintenance target device or a candidate for a diagnostic work for specifying a necessary work,
A maintenance support tree information management unit for managing maintenance support tree information in which the treatment work candidates and the diagnosis work candidates are configured in a tree shape;
Using the cost information of each work constituting the maintenance support tree information, an optimum work calculation unit for calculating the optimum work that minimizes the work cost or the work time;
A display unit for presenting the calculated optimum work;
With
The maintenance support tree information management unit updates the maintenance support tree information to reflect the result of the executed treatment work,
The optimum work calculation unit calculates the optimum work based on the updated maintenance support tree information. The maintenance support system according to claim 1,
In the maintenance support tree, the optimum work calculation unit uses the past number of cases, the experience value of the maintenance worker, and the information of the determination certainty that the result determined as Yes or No in the diagnosis work is a correct probability, A maintenance support system characterized by calculating a work start point at which an expected value of work cost or work time is minimized. The maintenance support system according to claim 1,
In order to minimize the expected value of work cost or work time under the condition that the maintenance worker assumes that the calculated optimum work is executed at the beginning of the maintenance work, it should be performed after the optimum work. An assumed work simulation unit for calculating two or more assumed works;
When the optimum work or the assumed work calculated immediately before is the treatment work at the lowest layer, the assumed work simulation unit excludes each of those works and the diagnosis work of the next higher hierarchy from the candidates for the assumed work calculation. Or, if it is not the lowest-level treatment work, compare the case points N _Y and N _N of the next lower layer, and assume the work or treatment work of the basic configuration block on the side where the case point value is smaller A maintenance support system, which performs a process of updating maintenance support Tree information excluded from calculation candidates. The maintenance support system according to claim 1,
A maintenance support system, further comprising a reception information management unit that statistically calculates a determination certainty factor from a result of failure diagnosis work performed in the past. A maintenance support program for supporting maintenance work by presenting candidates for treatment work such as parts replacement and adjustment necessary for maintenance work of manufacturing equipment and inspection equipment, or diagnostic work for specifying those treatment work,
Maintenance support tree information configured in a tree shape by adding connection information before and after the candidate treatment work necessary for recovery to each malfunction phenomenon of the maintenance target device and diagnostic work information for identifying them. The steps to manage
Included in alarm information issued to the support center or a report from the user based on an abnormal value indicated by data acquired from each sensor installed in the maintenance target device when a failure occurs in the maintenance target device Selecting the maintenance support tree corresponding to them from keywords;
During the maintenance support tree above, use the past case count, experience values input by maintenance workers, and cost information for diagnostic work and treatment work, and use the starting point of work where the expected value of work cost or work time is minimized. A step of calculating as the optimal work;
The maintenance support tree information is reflected to reflect the result of the maintenance work performed until the maintenance work is supported by presenting the optimum work on the display terminal module and the maintenance work of the maintenance worker is completed. Updating and repeating the calculation of the optimum work,
A maintenance support program characterized by causing a computer to execute. In the maintenance support program according to claim 5,
During the maintenance support tree, the number of past cases, the experience value of the maintenance worker, and the information of the determination confidence that the result determined to be Yes or No in the diagnosis work is the correct probability, and the expected work cost or work time. A maintenance support program, further causing a computer to execute a step of calculating a starting point of an operation having a minimum value. In the maintenance support program according to claim 5,
In order to minimize the expected value of work cost or work time under the condition that the maintenance worker assumes that the calculated optimum work is executed at the beginning of the maintenance work, it should be performed after the optimum work. Calculating at least two assumed tasks;
If the above-mentioned optimum work or the assumed work calculated immediately before is the lowest-level treatment work, the respective work and the diagnostic work of the hierarchy one level above are excluded from the candidates for assumed work calculation, or the lowest-level treatment work If it is not, maintenance is performed by comparing the case points N _Y and N _N of the next lower level and excluding the work of the basic configuration block or the treatment work on the side where the case point value is smaller from the candidates for the assumed work calculation A maintenance support program, further causing a computer to execute a step of updating support tree information. In the maintenance support program according to claim 5,
A maintenance support program that causes a computer to further execute a step of statistically calculating a determination certainty factor from a result of a fault diagnosis work performed in the past. A maintenance support device for supporting maintenance work by presenting candidates for treatment work such as parts replacement and adjustment necessary for maintenance work of manufacturing equipment and inspection equipment or diagnostic work for specifying those treatment work,
Maintenance support tree information configured in a tree shape by adding connection information before and after the candidate treatment work necessary for recovery to each malfunction phenomenon of the maintenance target device and diagnostic work information for identifying them. Maintenance support Tree information management means for managing
Included in alarm information issued to the support center or a report from the user based on an abnormal value indicated by data acquired from each sensor installed in the maintenance target device when a failure occurs in the maintenance target device Reception information management means for selecting the maintenance support tree corresponding to them from keywords,
In the maintenance support tree, the cost information for each work, the number of past cases, the experience value input by the maintenance worker, and the information of the determination certainty that the result determined as Yes or No in the diagnosis work is the correct probability. Use, an optimal work calculation means for calculating the starting point of the work where the expected value of work cost or work time is minimized,
In order to minimize the expected value of work cost or work time under the condition that the maintenance worker assumes that the calculated optimum work is executed at the beginning of the maintenance work, it should be performed after the optimum work. An assumed work simulation means for calculating two or more assumed works;
Supporting maintenance work by presenting the optimal and assumed work on the display terminal module, and reflecting the results of the maintenance work performed until the maintenance target device is restored. A diagnostic work execution means for updating the tree information and repeating the calculation of the optimum work and the assumed work;
The maintenance support tree information is updated while statistically generating the temporary work result of the optimum work and the subsequent assumed work to be used for calculating the assumed work from the number of past cases and the experience value of the maintenance worker. Maintenance support tree simulation information updating means for performing,
A determination certainty factor generating means for statistically calculating a determination certainty factor from the result of the fault diagnosis work performed in the past;
A case count unit that automatically calculates a count coefficient for adjusting the difference in weight per case of the number of cases or experience of maintenance workers caused by the difference in the number of treatment work hanging on each branch of the maintenance support tree Balancing means;
A maintenance support device comprising: The reception information management means further calculates the determination certainty statistically from the result of the fault diagnosis work performed in the past,
In the case where the above-mentioned optimal work or the assumed work calculated immediately before is the lowest level treatment work, the assumed work simulation means further selects each of those works and the diagnosis work of the next higher hierarchy from the candidates for the assumed work calculation. If it is excluded or it is not the lowest-level treatment operation, the case points N _Y and N _N of the next lower layer are compared, and the operation of the basic component block or the treatment operation on the side with the smaller case point value is performed. Update the maintenance support tree information to be excluded from the assumed work calculation candidates.
The maintenance support apparatus according to claim 9.

Images