Repair assistance system and repair assistance method

Abstract

has: a probability calculation result database including a combination countermeasure table and a probability, the combination countermeasure table storing in association with an error code and a multiple error code each corresponding to an error code indicating how many times the error code has occurred in the past, a countermeasure content corresponding to each of the error code and the multiple error code, the probability being a probability of obtaining the countermeasure content corresponding to each of the error code and the multiple error code, obtained from a predetermined calculation formula; a data processing section that performs combination generation processing of generating a combination table including a new error code and a new multiple error code based on device error information obtained from a device to be repaired, and repair prediction processing of predicting a recommended countermeasure content for the new error code and the new multiple error code based on the combination table and the probability calculation result database; and a result processing section that sorts and prompts a result of the repair prediction processing to a user.

Description

Technical Field

[0001] This invention relates to repair assistance systems and repair assistance methods. Background Technology

[0002] Conventionally, various techniques exist to assist in responding to malfunctions occurring in devices and equipment. For example, in Patent Document 1, an extraction unit determines the format of each error log obtained from the monitored device and extracts each determined format as a log pattern. Then, a generation unit determines which log pattern matches each error log included in a day and generates a log pattern list composed of log patterns. Then, a calculation unit calculates the similarity between the error log at the time of the malfunction and the log pattern list of the event date, and an output unit outputs a malfunction solution for the event inferred from the error log at the time of the malfunction based on the similarity.

[0003] Existing technical documents

[0004] Patent documents

[0005] Patent Document 1: Japanese Patent Application Publication No. 2019-49802 Summary of the Invention

[0006] The problem that the invention aims to solve

[0007] In Patent Document 1, a fault solution is output based on the similarity between the error log at the time of the fault and the error log on the date of the incident, inferred from the error log at the time of the fault. However, in devices and equipment with complex internal mechanisms, such as ATMs (Automated Teller Machines), it is sometimes difficult to determine the true location of the fault simply by relying on the similarity of the error logs. For example, although an error is output from the banknote storage section of the ATM, if the banknotes being transported from that section are actually not properly conveyed along the transport path, or if the poor transport in a bidirectional transport path is caused by a switching gate, it is difficult to infer the repair location when the error is output from a location other than the actual fault location. In such cases, it is necessary to rely on the experience and skills of engineers, and when engineers with low experience and skills are responsible for repairing such devices and equipment, unnecessary maintenance costs are incurred.

[0008] One objective of this invention is to provide a repair assistance system and repair assistance method capable of providing highly accurate indications of diagnostic materials for determining the true location of a fault.

[0009] Methods for solving problems

[0010] One aspect of the repair assistance system of the present invention includes: a probability calculation result database, comprising: a combination response table, which stores error code groups containing error codes of the device to be repaired and multiple error codes indicating that the error codes have occurred multiple times in the past, and corresponding response content for each of the error codes and the multiple error codes; and a probability, which is the probability of obtaining the error codes and the corresponding response content for each of the multiple error codes obtained from a predetermined formula; a data processing unit that performs combination generation processing and repair prediction processing, wherein the combination generation processing generates a combination table containing new error codes and new multiple error codes based on device error information obtained from the device to be repaired, and the repair prediction processing predicts recommended response content for the new error codes and the new multiple error codes based on the combination table and the probability calculation result database; and a result processing unit that sorts the results of the repair prediction processing and displays them to the user.

[0011] Invention Effects

[0012] According to one aspect of the present invention, it is possible to provide highly accurate information on the judgment material used to determine the true location of the fault. Attached Figure Description

[0013] Figure 1 This is a block diagram illustrating the functional structure of the repair assistance system in this embodiment.

[0014] Figure 2 It is a schematic diagram showing the hardware structure of a computer.

[0015] Figure 3 This is a diagram illustrating an example of a flowchart of pre-processing data storage / computation processing performed by a pre-processing server.

[0016] Figure 4 This is a diagram illustrating an example of device error history information.

[0017] Figure 5 This is a diagram showing examples of the conditional information and background information contained in the device's background information.

[0018] Figure 6 This is a diagram illustrating an example of correct repair response information.

[0019] Figure 7 This is a diagram showing an example of device error history information after code conversion.

[0020] Figure 8 This is a diagram illustrating an example of a comprehensive code table redefined by the data processing department.

[0021] Figure 9 This is a diagram illustrating an example of a combined response table.

[0022] Figure 10 This is a diagram representing an example of the probability calculation result DB.

[0023] Figure 11 This is a diagram illustrating an example of determining repair information.

[0024] Figure 12 This is a diagram illustrating an example of a flowchart of the prediction process performed by a prediction server.

[0025] Figure 13 This is a diagram illustrating an example of a device error message obtained by the data acquisition unit of the prediction server.

[0026] Figure 14 This is a diagram showing an example of a combination table.

[0027] Figure 15 This is a diagram illustrating an example of a repair prediction result.

[0028] Figure 16 This is a diagram illustrating an example of recommended repair information. Detailed Implementation

[0029] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. The following description and drawings are illustrative of the invention, and appropriate omissions and simplifications have been made for clarity. The present invention can also be implemented in various other ways. Unless otherwise specified, each constituent element may be a single element or a plurality of elements.

[0030] To facilitate understanding of the invention, the positions, sizes, shapes, and extents of the constituent elements shown in the accompanying drawings may not represent their actual positions, sizes, shapes, or extents. Therefore, the invention is not necessarily limited to the positions, sizes, shapes, and extents disclosed in the accompanying drawings.

[0031] In the following descriptions, various information is sometimes represented using terms like "table" or "list," but this information can also be represented using other data structures. To indicate independence from data structures, terms like "XX table" or "XX list" are sometimes referred to as "XX information." When describing identification information, terms such as "identification information," "identifier," "name," "ID," and "number" are interchangeable.

[0032] When multiple constituent elements have the same or identical functions, different subscripts are sometimes used to describe the same symbol. However, when it is not necessary to distinguish these multiple constituent elements, the subscripts are sometimes omitted for description.

[0033] Furthermore, in the following description, the processing performed by executing a program is sometimes described, but the program is executed by a processor (e.g., CPU (Central Processing Unit), GPU (Graphics Processing Unit)), which appropriately uses storage resources (e.g., memory) and / or interface devices (e.g., communication ports) to perform the determined processing; therefore, the main body of the processing can also be the processor. Similarly, the main body of the processing performed by executing a program can also be a controller, device, system, computer, or node having a processor. The main body of the processing performed by executing a program can be any arithmetic unit, but it can also include dedicated circuitry (e.g., FPGA (Field-Programmable Gate Array), ASIC (Application Specific Integrated Circuit)) that performs specific processing.

[0034] The program can also be installed onto a device such as a computer from a program source. The program source can be, for example, a program distribution server or a computer-readable storage medium. When the program source is a program distribution server, the program distribution server can also include a processor and storage resources for storing the programs to be distributed, and the processor of the program distribution server distributes the programs to other computers. Furthermore, in the following description, two or more programs can be implemented as one program, or one program can be implemented as two or more programs.

[0035] The following describes in detail the application of the repair assistance system and repair assistance method of this embodiment to an ATM, but it is not limited to this example and can be applied to various devices and equipment.

[0036] Figure 1 This is a block diagram illustrating a functional structure example of the repair assistance system 1000 in this embodiment. For example... Figure 1 As shown, the repair assistance system 1000 is configured to have a preprocessing server 400 and a prediction server 500.

[0037] The preprocessing server 400 includes: a data acquisition unit 401 that receives error logs and repair / replacement component data output from the repair target device 100 from the data transceiver device 200; an error log DB 402 that stores the data output by the data acquisition unit 401; basic data DB 403 that is input from a basic data input device 300 that inputs basic data related to the repair target device 100; a data processing unit 404 that uses the error log 402 and the basic data DB 403 to process and calculate the data stored in the probability calculation result DB 405; and the probability calculation result DB 405 that stores the processing result of the data processing unit 404.

[0038] The prediction server 500 includes: a data acquisition unit 501 that receives error logs from the repair target device 100 from the data transceiver device 200; an error log DB 502 that stores data output by the data acquisition unit 501; basic data DB 503 that is the same as the basic data DB 403 possessed by the preprocessing server 400; a data processing unit 504 that processes and calculates data for reference probability calculation result DB 505 using the error log 502 and the basic data DB 503; a probability calculation result DB 505 that is referenced using the processing result of the data processing unit 504; an extraction result processing unit 506 that extracts recommended repair information to be suggested to the user using the reference result of the probability calculation result DB 505; and a data transmission unit 507 that sends the recommended repair information extracted by the extraction result processing unit 506 to the data transceiver device 200. Additionally... Figure 1 The arrows in the diagram indicate the direction of data flow. For recommended repair information, use... Figure 16 The details will be provided later.

[0039] The servers shown above, for example, can be accessed through... Figure 2 A general computer 200, as shown in the schematic diagram of a computer, includes a CPU 201, a memory 202, an external storage device 203 such as an HDD (Hard Disk Drive), a read / write device 207 for reading and writing information on removable storage media 208 such as CDs (Compact Disks) and DVDs (Digital Versatile Disks), an input device 206 such as a keyboard and mouse, an output device 205 such as a display, a communication device 204 such as a NIC (Network Interface Card) for connecting to a communication network, and an internal communication line (called a system bus) 209 connecting them.

[0040] For example, error logs (DB), basic data (DB), probability calculation results (DB), and other databases stored in each server can be read from memory (202) or external storage device (203) and utilized by CPU 201. Furthermore, the data acquisition unit and data processing unit of each server, and the result extraction processing unit (506) and data transmission unit (507) of the prediction server 500 can be implemented by CPU 201 loading a predetermined program stored in external storage device 203 into memory (202) and executing it. Additionally, each server may have an input unit that enables input function by activating input device 206 via CPU 201. Furthermore, each server may have an output unit that enables output function by activating output device 205 via CPU 201. Additionally, each server may have a communication unit that enables communication function by activating communication device 204 via CPU 201. In this embodiment, it is assumed that the data acquisition unit of each server and the data transmission unit of the prediction processing server 500 have the functions performed by the aforementioned communication unit.

[0041] The aforementioned predetermined program can also be stored (downloaded) from storage medium 208 via read / write device 207 or from network via communication device 204 to external storage device 203, and then loaded onto memory 202 for execution by CPU 201. Alternatively, it can be directly loaded onto memory 202 from storage medium 208 via read / write device 207 or from network via communication device 204 for execution by CPU 201.

[0042] The preprocessing server 400 and the prediction server 500 are described below as hardware units installed in a general computer. However, all or part of them can also be distributed across one or more computers in a cloud, communicating with each other to achieve the same function. Alternatively, a single server can be used to constitute both the preprocessing server 400 and the prediction server 500. Flowcharts are used to illustrate the actions of each unit of the preprocessing server 400 and the prediction server 500, and to provide examples of the data they hold.

[0043] Figure 3 This is a diagram illustrating an example of a flowchart of the pre-processing data storage / computation process performed by the pre-processing server 400. Hereinafter, it is assumed that this process occurs whenever the device 100 to be repaired malfunctions and is repaired.

[0044] The data acquisition unit 401 of the preprocessing server 400 acquires device error history information 4031, which is a history (log) of error information of the repair target device 100 or its components, from the repair target device 100 or the data transceiver device 200 at a time when the repair target device 100 malfunctions, at a time when the repair is performed, or at a time when the repair is completed, and stores it in the error log DB402 (S301). At any one or more time points before, simultaneously with, or after the repair case, the preprocessing server 400 acquires various data stored in the basic data DB403 from the basic data input device 300, and stores the acquired data in the basic data DB403 through the basic data input device 300. The various data include: device background information 4032, which contains background information of the device as a business insight related to the device being repaired, such as the installation location of the equipment 100 (e.g., Indonesia), the installation period (e.g., October 2010), operating performance, component replacement performance (e.g., no rubber replacement within the past six months), the manufacturing period of the components and the device (e.g., manufactured in 2010), and the model name; and condition information (described later) that determines the preconditions for the data processing unit 404 to perform calculations; and correct repair response information 4033, which indicates the actual response method when the error occurs (e.g., repair response for the rubber roller). The data processing unit 404 reads this information from the basic data DB403, namely, past error logs, year, month, region, operating time, and other fault cause data that are assumed to be the cause of the fault based on the correct repair data when the equipment being repaired malfunctions.

[0045] Figure 4 This is a diagram illustrating an example of device error history information 4031. (Example:) Figure 4 As shown, the device error history information 4031 stores the error code number (#), an example of an error code indicating an error that occurred in the repaired device 100 or its components, and an elapsed time from the occurrence of the error code to the present, corresponding to each repair case. The elapsed time is assumed to be a pre-calculated time in the repaired device 100 or its components.

[0046] exist Figure 4 For example, error code "xx" with number 1 indicates information from 1 hour ago in the error history information 4031 of the data acquisition unit 401, and error code "uu" with number 7 indicates information from 1200 hours ago in the error history information 4031 of the data acquisition unit 401.

[0047] Figure 5 This is a diagram illustrating examples of the conditional information and background information included in the device background information 4032. Figure 5In the example, as conditional information, the data includes a list of useless codes (40321), a data for a specified period (40322), and code conversion data (40323). Additionally, as background information, the data includes background information (40324).

[0048] The useless code list data 40321 is a list of error codes that are not needed for repair prediction among the error codes acquired by the data acquisition unit 401. When the device 100 to be repaired is a device with a complex structure, such as an ATM, error codes other than those for the repair target (e.g., a warning message indicating insufficient banknotes always stored in the storage compartment) should be output; therefore, such error codes are pre-registered as useless codes. Figure 5 In the example, it indicates that the error code "vv" is registered as a useless code.

[0049] The specified period data 40322 is data that records the period during which past error codes obtained by the data acquisition unit 401 that have passed a certain period are excluded from processing. Figure 5 The example shows a case where error codes older than 1000 hours are excluded from processing.

[0050] Code conversion data 40323 is a conversion table used to summarize error codes obtained by the data acquisition unit 401 that have similar error content. Figure 5 For example, it shows how to convert both error codes "xx" and "ww" into the comprehensive code "XX".

[0051] Background information data 40324 indicates the background and environment requiring repair, including information related to the equipment 100 and components to be repaired, as described above. Figure 5 The document shows cases where "zz" is registered as background and environmental information representing business insights related to the aforementioned repaired device.

[0052] In addition, the correct repair response information 4033 representing the response content in past repairs is stored in the basic data DB403. Figure 6 This is a diagram illustrating an example of the correct repair response information 4033. (See diagram below.) Figure 6 As shown, the correct repair response information 4033 stores the ID of past repair cases and the response content (in this case, the repaired component) in relation to those repairs. Figure 6 The diagram illustrates the situation where repair case "0001" was handled by repairing part "AA". Figure 6 The document provides examples of component repairs, but also stores maintenance information beyond repairs, such as component replacement, repair, or no replacement.

[0053] Return to Figure 3 In the preprocessing server 400, the data processing unit 404 performs preprocessing (S302) on a data set containing information for each repair case in the error log DB402 and the basic data DB403 as described above. Preprocessing refers to processing the acquired data to reduce and improve the accuracy of the probability calculation processing described later.

[0054] Specifically, the data processing unit 404 refers to the device error history information 4031 and the useless code list data 40321, and deletes the record containing the error code registered in the useless code list data 40321 from the device error history information 4031. For example, the data processing unit 404 deletes the record containing the error code registered in the useless code list data 40321 from the device error history information 4031. Figure 4 The device error history information 4031 shown deletes record number 5, which is the record of error code "vv" registered as a useless code in the useless code list data 40321.

[0055] Additionally, the data processing unit 404, referring to the device error history information 4031 after deleting useless codes and the specified period data 40322, deletes from the device error history information 4031 records containing elapsed times prior to the specified period registered in the specified period data 40322. For example, the data processing unit 404 deletes from... Figure 4 The device error history information 4031 shown deletes record number 7, which is the record of the elapsed time before the period "1000" hours that was recorded as the specified period in the specified period data 40322.

[0056] Furthermore, the data processing unit 404, referring to the device error history information 4031 after deleting useless codes and codes outside the specified period, and the code conversion data 40323, determines from the device error history information 4031 a record containing the error code registered in the code conversion data 40323, and converts the determined error code of that record into a composite code. For example, the data processing unit 404 from... Figure 4 In the device error history information 4031 shown, the record number 1, which contains the error code "xx" registered as an error code to be converted into code conversion data 40323, and the record number 3, which contains "ww", are identified. The error codes of these identified records are converted into the comprehensive code "XX".

[0057] After performing these useless code removals, out-of-specification code removals, and conversions to comprehensive codes, the device error history information 4031 becomes... Figure 7 The device error history information shown is in the state of 701. Figure 7As can be seen from the above processing, the comprehensive code converted from the error code is matched with the elapsed time, and records numbered 5 and 7 are deleted (—).

[0058] Subsequently, the data processing unit 404 sums up the number of integrated codes stored in the device error history information 4031 after deleting useless codes, deleting codes outside the specified period, and converting them to integrated codes, and generates an integrated code table that redefines multiple codes into new integrated codes. As described later, the integrated code table stores error code groups that include integrated codes and redefined integrated codes.

[0059] Figure 8 This is a diagram illustrating an example of the comprehensive code table redefined by the data processing unit 404. (See diagram below.) Figure 8 As shown, synthesis code table 801a stores the synthesis code number (#) and its corresponding synthesis code value. Figure 8 In (a), for example, a composite code table 801a is shown where the data processing unit 404 generates a composite code table 801a for error code "XX", which is redefined with the composite code "XX" indicating the occurrence of the code as number 1 and the composite code "XX multiple times" indicating the occurrence of the code multiple times as number 2. Figure 7 As can be seen, since records numbered 1, 3, 4, and 5 are error codes "XX", a comprehensive code table 801a is generated containing records for which the comprehensive code "XX times" has been redefined. In this example, error codes that appear at least twice are redefined as new codes. However, a threshold (N times) for the number of times a code appears can be determined based on factors such as the type of equipment and component being repaired, the usage environment of the equipment and component, and the installation environment. The data processing unit 404 redefines the code when it appears above this threshold. Furthermore, the data processing unit 404 can set multiple such thresholds based on the type of error code, and redefine the code when each error code appears above its respective threshold. Regarding these thresholds, by calculating the probability of a correct prediction in advance and setting the best condition as the threshold, the prediction accuracy can be improved.

[0060] Furthermore, the data processing unit 404 generates a comprehensive code table 801b that appends background information data 40324 to the generated comprehensive code table 801a. Figure 8 From (b), it can be seen that it was generated in Figure 8 In the comprehensive code table 801a shown in (a), “ZZ” was added to the background information data 40324 to form the comprehensive code table 801b with number 4.

[0061] Return to Figure 3When the processing up to S302 ends, the data processing unit 404 performs combination generation and probability calculation processing (S303).

[0062] Specifically, the data processing unit 404 generates a combination response table that corresponds to a combination of one or more comprehensive codes stored in the comprehensive code table 801b with the response content (e.g., correct repair of the part) when an error code is generated, which is the basis of the comprehensive code contained in the combination.

[0063] Figure 9 This is a diagram illustrating an example of a combined response table. In Figure 9 As an example, the example shows combined response tables 9011, 9012, and 9013 generated for each repair case when maintenance was performed on the equipment 100 at three different time points. That is, a combined response table is stored for each repair case. These tables can also be aggregated together.

[0064] like Figure 9 As shown, the combination response table includes the combination number (#), the synthesis code contained in the combination, and the response content when the error code that forms the basis of the synthesis code is generated (in...). Figure 9 The correct repair parts are stored together. This allows for tracking via the data processing unit 404. Figure 6 The correct repair response information 4033 shown indicates the correct repair component corresponding to the assembly. Figure 9 For example, the example shows a case where the repair part "AA" is stored for the combination "XX" and "XX multiple times". In addition, combinations of individual synthesis codes (e.g., "XX") and multiple synthesis codes (e.g., "XX multiple times") are not generated, and the swapping of synthesis codes is excluded (e.g., swapping the order of "YY, XX" in the group of "XX, YY").

[0065] Thus, when the data processing unit 404 generates a combination response table based on the repair case, it then performs probability calculation processing. For example, as part of the probability calculation processing, the data processing unit 404 calculates the support level for each combination stored in the combination response table 9011. For example, for the combination codes XX, YY and repair part AA, the support level can be calculated using the formula shown below.

[0066] [Formula 1]

[0067]

[0068] Based on the above formula, by calculating the support of each combination, it can be used as a criterion for determining whether a combination is a frequently occurring error code.

[0069] Furthermore, for example, as a probability calculation process, the data processing unit 404 calculates the confidence of each combination stored in the combination response table 9011. For example, for the combination codes XX, YY and repair parts AA, the confidence can be calculated using the formula shown below.

[0070] [Formula 2]

[0071]

[0072] Based on the above formula, by calculating the confidence level of each combination, the probability of a specific repair occurring in a certain error code can be calculated.

[0073] Additionally, as a probability calculation process, the data processing unit 404 calculates the lift value for each combination stored in the combination response table 9011. For example, for combination codes XX, YY and repair parts AA, the lift value can be calculated using the formula shown below.

[0074] [Formula 3]

[0075]

[0076] By using the above formula to calculate the boost value for each combination, repairs that occur frequently and are unrelated to the log can be eliminated. In this example, actions with a boost value below 1 are not recommended.

[0077] When performing these probability calculations, the data processing unit 404 performs a lightweighting process (S304) to exclude combinations that do not meet predetermined thresholds. For example, the data processing unit 404 excludes combinations whose boost value is less than the threshold "1". Furthermore, for example, the data processing unit 404 excludes combinations whose support is less than the threshold "0.1".

[0078] The data processing unit 404 stores the data (S305) that corresponds to the combination response table, the number of combinations, the support, confidence and boost value of each combination, excluding the combinations excluded in the lightweight processing, as the probability calculation result DB405.

[0079] Figure 10 This is a diagram representing an example of the probability calculation result DB405. (Example...) Figure 10 As shown, the probability calculation result DB405 will include the combination number (#), the synthesis code contained in the combination, and the response content when the error code that forms the basis of the synthesis code is generated (in...). Figure 10 The correct repair parts are stored together. Figure 10For example, the example shows a case where the support for repair part "BB" with the combination "XX times" is "0.9", the confidence level is "90", and the boost value is "10". Additionally, the combination "XX times" indicates that the synthesis code that forms the basis appears more than twice.

[0080] When the processing of S305 is finished, Figure 3 The pre-processing server 400 shown completes the pre-data storage / computation processing. In the example above, if a response to an error code or a composite error code has been determined, the basic data input device 300 or the data acquisition unit 401 may pre-store its contents in the basic data DB 403.

[0081] Figure 11 This is a diagram illustrating an example of specific repair information stored containing information that determines the response to a specific error code or a combination of error codes. For example... Figure 11 As shown, the repair information determination 1101 stores the determination information, which indicates the type of response (e.g., repair) required, the error code or combined error code, the recommended response (e.g., recommended repair) indicating the response to the determination information, and the reason for requiring the response. Figure 11 For example, the recommended repair for an error code with the specific information "ZZ" or a comprehensive error code is "CC" (e.g., replacement), with the reason stated as "poor design".

[0082] Next, the prediction processing performed by prediction server 500 will be explained. Figure 12 This is a diagram illustrating an example of a flowchart of the predictive processing performed by the predictive server 500. Hereinafter, it is assumed that this process is performed whenever a new fault occurs in the device 100 to be repaired.

[0083] When a new fault occurs in the device 100 under repair, or when repair is initiated, the data acquisition unit 501 of the prediction server 500 acquires device error information 5021, which represents the history (log) of error information of the device 100 under repair and its components, from the device 100 under repair or the data transceiver 200, and stores it in error log DB502 (S1201). At this time, the data acquisition unit 501 of the prediction server 500 reads various data contained in the same basic data DB503 as the basic data DB403 created in the preprocessing server 400. For example, the data acquisition unit 501 reads device background information 5031, which includes background information such as the installation location, installation period, operating performance, component replacement performance, manufacturing period or model name of the component or device, and condition information that determines the prerequisites for calculation by the data processing unit 504, as various data.

[0084] Figure 13 This diagram illustrates an example of device error information 5021 acquired by the data acquisition unit 501 of the prediction server 500. The device error information 5021 stores the time of the new fault occurrence. Figure 4 The device error history information 4031 shown contains the same items. These items are the same as... Figure 4 Since they are the same, their description is omitted here.

[0085] Then, the data processing unit 504, similar to the data processing unit 404 of the preprocessing server 400, reads the error log DB502 and the device background information 5031 and performs preprocessing (S1202). When the processing in S1202 is completed, the data processing unit 504 performs combined generation (S1203). Specifically, the data processing unit 504 targets... Figure 13 The device error message 5021 at the time of the new fault occurrence is shown, and a combination table is generated that stores one or more combinations of the comprehensive codes stored in the comprehensive code table 801b, which is the same as the comprehensive code table generated by the data processing unit 404 of the preprocessing server 400.

[0086] Figure 14 This is a diagram representing an example of combination table 1401. For example... Figure 14 As shown, the combination table stores the combination number (#) and the comprehensive code contained within that combination. These items and... Figure 9 Since they are the same, their description is omitted here.

[0087] Subsequently, the data processing unit 504 reads the same probability calculation result DB505 as the probability calculation result DB405 generated by the data processing unit 404 of the preprocessing server 400 (S1204). It then interfaces the read probability calculation result DB505 with the combination table 1401 generated in S1203, extracting the comprehensive code corresponding to each combination contained in the combination table 1401, the response content when an error code is generated (e.g., correct repair of the component), the recommended repair, and the confidence level and improvement value. This is output as a repair prediction result (S1205). Support may also be included in this repair prediction result.

[0088] Figure 15 This is a graph representing an example of repair prediction result 1501. (Example) Figure 15 As shown, the repair prediction result 1501 stores the combinations in combination table 1401, the corresponding responses (i.e., recommended repairs) stored in the probability calculation result DB505, and the associated confidence and improvement values. Figure 15For example, as a response to the combination "XX times", "BB" (e.g., component replacement) is recommended, indicating a credibility of 90% and an improvement value of 10 points. Figure 15 In the middle, the data processing unit 504 rearranges the records in order of confidence from high to low, and outputs the repair prediction result 1501.

[0089] In addition, the data processing unit 504 reads the confirmed repair information 1101 stored in the preprocessing server 400 from the basic data DB503. Figure 11 Similarly, the repair information is determined (S1206), and the determination information, recommended repair, and reason are extracted, which include the comprehensive code corresponding to each combination contained in the combination table 1401.

[0090] The result extraction and processing unit 506 combines the confirmed repair information extracted from the basic data DB503 with the repair prediction result 1501 to determine the recommended repair order to be displayed as the final recommended repair information to be presented to the user (S1208). Furthermore, the result extraction and processing unit 506 outputs the recommended repair information in the determined order to the data transmission unit 507, which then presents the recommended repair information to the user (S1209). As a method for presenting the recommended repair information, for example, the data transmission unit 507 sends the aforementioned recommended repair information to the data transceiver device 200, and the display unit of the data transceiver device 200 displays the recommended repair information.

[0091] Figure 16 This is a diagram illustrating an example of recommended repair information 1601 after the result extraction and processing unit 506 has determined the order. (See diagram for example.) Figure 16 As shown, in the recommended repair information, the priority order, recommended repair, confidence level, improvement value, and basis for the recommended repair are stored together. When the result extraction processing unit 506 extracts confirmed repair information, it determines that the response is necessary and rearranges it to be at the top of the recommended repair information, setting the confidence level and improvement value to indicate that they are necessary (e.g., required). Furthermore, the result extraction processing unit 506 sets the reasons for determining the repair information based on these criteria. The result extraction processing unit 506 then rearranges the repair prediction results 1501 according to confidence level and sets the combination based on these criteria.

[0092] exist Figure 16In the data, it can be seen that in the highest priority order "1", the recommended repair "CC" is presented, indicating that the reliability and improvement value of this recommended repair are both indicated as "necessary", and the basis for these recommendations is "poor design". Furthermore, in the next priority order "2", the recommended repair "BB" with the highest reliability among the repair prediction results 1501 is presented, indicating that the reliability and improvement value of this recommended repair are both indicated as "90%" and "10 points", respectively, based on the indication that this is caused by the occurrence of the comprehensive error code "XX times". By presenting such recommended repair information 1601 to the user, the user can easily determine the actual location of the fault based on the presented information.

[0093] Thus, according to this embodiment, due to the presence of the following components, it is possible to provide highly accurate judgment materials for determining the true location of the fault: a probability calculation result database (e.g., probability calculation result DB405), which includes a combination response table (e.g., combination response table 9011) and probabilities (e.g., support, confidence, enhancement value). The combination response table stores error code groups containing the error code of the repair target device 100 (e.g., error code "XX") and multiple error codes indicating that the error code has occurred multiple times in the past (e.g., "XX multiple times"), along with corresponding response content for each error code and multiple error code. The probabilities are calculated from predetermined formulas (e.g., formulas 1 to 3). The system obtains the probabilities of corresponding responses for each error code and multiple error codes from various sources; a data processing unit (e.g., data processing unit 504) performs combination generation processing and repair prediction processing. The combination generation processing generates a combination table (e.g., combination table 1401) containing new error codes and new multiple error codes based on device error information (e.g., device error information 5021) obtained from the device 100 to be repaired; the repair prediction processing predicts recommended responses for new error codes and new multiple error codes based on the combination table and the probability calculation result database; and a result processing unit (e.g., data processing unit 505) sorts the results of the repair prediction processing and displays them to the user.

[0094] Furthermore, the data acquisition unit (e.g., data acquisition unit 401) that obtains past log information (e.g., device error history information 4031) containing error codes from the repair target device 100, and the data processing unit counts the number of occurrences of each error code contained in the log information, and generates the probability calculation result database containing the error code group that defines the error codes that are counted multiple times as a multiple error code and stores the corresponding response content, so that the data used in the repair prediction process can be stored in advance.

[0095] In addition, the data acquisition unit obtains business insights, i.e., background information of the device, related to the device being repaired from the device being repaired or other devices. The data processing unit generates the combination response table, which includes the error code and the combination of each of the multiple error codes with the background information. Therefore, based on the business insights of the device being repaired, the user can provide information for repair and other responses.

[0096] Furthermore, as a result of the aforementioned repair prediction processing, the result processing unit outputs probability calculation result information to the display unit. This probability calculation result information includes the recommended response, the probability of taking the recommended response, and the error code or multiple error codes indicating the basis for recommending the recommended response. Therefore, it is possible to determine the appropriate response at a glance.

[0097] The data processing unit reads the determined repair information from the storage unit. This determined repair information is information that determines the response to the error code or the combined error code. It includes the recommended response, information indicating that the response is necessary, and the basis for the necessity of the response. The result processing unit outputs the probability calculation result information and the determined repair information as the result of the repair prediction processing to the display unit. Therefore, when a necessary response exists, the user can easily understand that the response is required even if the user has no understanding.

[0098] In the past, with devices and equipment like ATMs that have complex internal mechanisms, it was sometimes difficult to pinpoint the location of a fault simply by relying on the similarity of error logs. For example, if a banknote is bent due to a malfunction at part A, and an error occurs due to banknote congestion at part B (different from part A), simply addressing the error at part B cannot eliminate the malfunction at part A. In such cases, depending on the engineer's experience and skills, it might be necessary to address part A by replacing components in part A that did not produce the error. However, by storing this approach in relation to the current error (in this case, the error at part A), even engineers with limited experience and skills can easily identify the true location of the fault, thus reducing maintenance costs.

[0099] Furthermore, by identifying features from recurring error codes in logs that were previously unacceptable, rules for repair can be extracted with high precision, enabling the issuance of appropriate repair instructions. This technology can also be applied to basket analysis, which analyzes buyer characteristics. For example, previously, regardless of the customer type, products were recommended to customers who purchased one or more cans of beer (corresponding to multiple error codes). However, with this system, tea can be recommended to customers who purchased 12 or more cans of beer (corresponding to multiple error codes). This allows for more accurate product recommendations to customers making concentrated purchases.

[0100] Symbol Explanation

[0101] 1000 Repair Assist System

[0102] 400 Preprocessing Server

[0103] 401 Data Acquisition Department

[0104] 402 Error Log DB

[0105] 403 Basic Data Database

[0106] 404 Data Processing Department

[0107] 405 Probability Calculation Results (DB)

[0108] 500 prediction server

[0109] 501 Data Acquisition Department

[0110] 502 Error Log DB

[0111] 503 Basic Data Database

[0112] 504 Data Processing Department

[0113] 505 Probability Calculation Result DB

[0114] 506 Extraction Result Processing Department

[0115] 507 Data Transmission Department.

Claims

1. A repair assistance system, characterized in that, have: The data acquisition unit obtains past log information containing error codes from the device being repaired; The probability calculation result database includes: a combined response table, which stores error code groups containing error codes of the device to be repaired and multiple error codes indicating that the error code has occurred multiple times in the past, and corresponding response contents for the error codes and the multiple error codes respectively; and a probability, which is the probability of obtaining the error code and the corresponding response contents for the multiple error codes, obtained from a predetermined formula. The data processing unit performs a combination generation process and a repair prediction process. The combination generation process generates a combination table containing new error codes and new multiple error codes based on device error information obtained from the device to be repaired. The repair prediction process predicts recommended responses for the new error codes and new multiple error codes based on the combination table and the probability calculation result database. The results processing unit sorts the results of the repair prediction process and presents them to the user.

2. The repair assistance system according to claim 1, characterized in that, The data processing unit counts the number of times each error code in the log information occurs, and generates a probability calculation result database containing a combined response table that stores the error code group (where each error code is defined as a multiple error code) and the corresponding response content.

3. The repair assistance system according to claim 2, characterized in that, The data acquisition unit obtains business insights, i.e., background information of the device, related to the device being repaired or other devices. The data processing unit generates a combination response table containing the error codes and each combination of the multiple error codes with the background information.

4. The repair assistance system according to claim 1, characterized in that, The result processing unit outputs the probability calculation result information to the display unit as the result of the repair prediction processing. The probability calculation result information includes the recommended response, the probability of taking the response, and the error code or the multiple error codes indicating the basis for recommending the response.

5. The repair assistance system according to claim 4, characterized in that, The data processing unit reads repair information from the storage unit. This repair information determines the appropriate response to the error code or multiple error codes, and includes the recommended response, information indicating that the response is necessary, and the basis for this necessity. The result processing unit outputs the probability calculation result information and the determined repair information as the result of the repair prediction processing to the display unit.

6. A computer-aided repair method, characterized in that, The repair assistance method performs the following processing: Obtain past log information containing error codes from the device being repaired; Read out the probability calculation result database, which contains a combined response table, which stores error code groups containing the error code of the device to be repaired and multiple error codes indicating that the error code has occurred multiple times in the past, and the corresponding response content of the error code and each of the multiple error codes; And the probability, which is obtained from a predetermined formula, is the probability of obtaining the error code and the corresponding response content for each of the multiple error codes; Based on the device error information obtained from the device being repaired, a combination table containing the new error code and the new multiple error codes is generated; Based on the combination table and the probability calculation result database, predict the recommended response for the new error code and the new multiple error codes; as well as The predicted results are sorted and presented to the user.

7. The repair auxiliary method according to claim 6, characterized in that, The number of occurrences of each error code contained in the log information is counted, and a probability calculation result database is generated, which stores the error code group (defined as a group of multiple error codes) and the corresponding response content.

8. The repair auxiliary method according to claim 7, characterized in that, Obtain business insights, i.e., background information about the device, related to the device being repaired or other devices. Generate the combined response table, which includes the error code and each combination of the multiple error codes with the background information.

9. The repair auxiliary method according to claim 6, characterized in that, The probability calculation result information is output to the display unit as the prediction result, wherein the probability calculation result information includes the recommended response content, the probability of taking the response content, and the error code or the multiple error codes representing the basis for recommending the response content.

10. The repair auxiliary method according to claim 9, characterized in that, Repair information is read from the storage unit, wherein the repair information determines the response to the error code or the multiple error codes, and includes the recommended response, information indicating that the response is necessary, and the basis for the necessity of the response. The probability calculation result and the determined repair information are output to the display unit as the result of the prediction.

Images