METHOD FOR SUPPORTING THE MAINTENANCE TASK OF A PAPER SHEET PROCESSING SYSTEM

Abstract

METHOD FOR SUPPORTING MAINTENANCE TASK OF PAPER SHEET PROCESSING SYSTEM According to one embodiment, a method for supporting a maintenance task of a paper sheet processing system (11) comprises receiving a voice or speech signal regarding a request for a maintenance task transmitted from a first communication terminal (12) inside a workroom using short-distance wireless communication using a second communication terminal (13) outside the workroom, and performing voice recognition of the received voice or speech signal, extracting a keyword from the request for a maintenance task acquired by the voice recognition, and performing a maintenance support suggestion to retrieve the maintenance task content based on the extracted keyword,converting the result of retrieving the maintenance task content into a speech signal and transmitting the speech signal from the second communication terminal (13) to the first communication terminal (12) by short-distance wireless communication. Figure for abstract: Fig. 1.,

Description

Title of the invention: METHOD FOR SUPPORTING PAPER SHEET PROCESSING SYSTEM MAINTENANCE TASK Technical field

[0001] The embodiments described herein generally relate to a method of supporting a maintenance task of a paper sheet processing system.

[0002] Context

[0003] In a cash processing center of a bank, the introduction of communication devices such as a mobile phone, a mobile terminal, and the like from outside into the room is prohibited. In the maintenance service of a paper sheet processing system to be used in such a high-security environment, with regard to the emergency response of which immediate maintenance action to be performed in a short time is required, the maintenance task is performed by cooperation between a maintenance person at work working inside the room and a maintenance person on standby working outside the room via voice communication based on portable short-distance wireless communication devices (mobile terminals such as wireless transceivers).

[0004] As a working procedure of the maintenance task, the maintenance person at work inside the room reports the current state of the system to the maintenance person on standby outside the room (maintenance person on standby) by voice by means of the portable short-distance wireless communication device. The maintenance person on standby accesses the maintenance support system in which the past response examples (maintenance action) are managed, performs keyword retrieval to thereby confirm the contents of the responses, and informs the maintenance person at work inside the room of the contents of the responses.

[0005] As described above, in the conventional maintenance task, although the maintenance person on standby inputs a keyword for each category into the maintenance support system by using keywords obtained by using the portable short-range wireless communication device to thereby perform retrieval, there is sometimes a case in which the category to be input has some obscure parts and it takes a long time to manually perform retrieval to identify the contents. Furthermore, when the degree The operating skill of the maintenance person on standby regarding the maintenance support system is low, if similar recovery results are listed, the differences between them cannot be determined in an instant. In addition, when the response (maintenance action) is unclear (when recovery fails), it is necessary to ask a specialized analyst to solve the problem, and there is sometimes a case in which it takes a long time to obtain the determination.

[0006] As described above, in the conventional maintenance task support method of a paper sheet processing system used in the room of a cash processing center, an appropriate keyword is acquired from the content of a dialog between the inside of the work room and the outside of the work room to be performed by means of portable short-distance wireless communication devices, the response item is retrieved by using the maintenance support system, and the retrieved response item is then confirmed, and thus it takes a long time to inform the maintenance person inside the work room of the response task and cause the maintenance person inside the work room to execute the task.Furthermore, depending on the skill level of the maintenance person using the maintenance support system outside the workroom, a problem existed in that the recovery took a long time or it could not be instantly determined which of the similar recovery results is the response example (having a high degree of influence on the operation) in the case of the occurrence of an emergency.

[0007] This embodiment has been proposed to solve the above-mentioned problem, and aims to provide a method for supporting the maintenance task of a paper sheet processing system capable of supporting the maintenance task of a paper sheet processing system to be used inside a room of a cash processing center in the attempt that the maintenance task is executed quickly by automatically performing the maintenance task outside the working room.

[0008] The invention will be better understood and its advantages will be better understood upon reading the detailed description which follows. The description refers to the drawings indicated below and which are given as examples. Brief description of the drawings

[0009] [Fig. 1] [Fig. 1] is a block diagram showing the configuration of a maintenance management system to which the maintenance task support method according to one embodiment is applied.

[0010] [Fig.2] [Fig.2] is a block diagram showing the specific configuration of a voice information processing device used in the system shown in [Fig.l].

[0011] [Fig.3] [Fig.3] is a flowchart showing the flow of a series of processing of the voice information processing device shown in [Fig.2].

[0012] [Fig.4] [Fig.4] is a view for explaining the maintenance knowledge recovery processing of the maintenance support system to be used in the system shown in [Fig.l].

[0013] [Fig.5] [Fig.5] is a view illustrating the recording contents of a knowledge database to be used in the system shown in [Fig.l].

[0014] [Fig.6A] [Fig.6A] is a view showing an example of a first part for explaining the calculation of the association degree and the processing rate from the keyword retrieval result, the rearrangement of the order of the association degree, the evaluation, and the processing rate, and the provision of information regarding the final result in the system shown in [Fig.1].

[0015] [Fig.6B] [Fig.6B] is a view showing an example of a second part for explaining the calculation of the association degree and the processing rate from the keyword retrieval result, the rearrangement of the order of the association degree, the evaluation, and the processing rate, and the provision of information regarding the final result in the system shown in [Fig.l].

[0016] [Fig.6C] [Fig.6C] is a view showing an example of a third part for explaining the calculation of the association degree and the processing rate from the keyword retrieval result, the rearrangement of the order of the association degree, the evaluation, and the processing rate, and the provision of information regarding the final result in the system shown in [Fig.l].

[0017] [Fig.6D] [Fig.6D] is a view showing an example of a fourth part for explaining the calculation of the association degree and the processing rate from the keyword retrieval result, the rearrangement of the order of the association degree, the evaluation, and the processing rate, and the provision of information regarding the final result in the system shown in [Fig.l].

[0018] [Fig.7] [Fig.7] is a view showing examples of questions or requests for confirmations to be asked or made by the maintenance person at work when the knowledge number of the knowledge database shown in Fig. 6 is to be selected. Description of the embodiments

[0019] Hereinafter, one embodiment will be described with reference to the accompanying drawings.

[0020] Generally, according to one embodiment, a method for supporting the maintenance task of a paper sheet processing system configured to support a task of a paper sheet processing system used inside a workroom of a cash processing center of a bank workroom comprises receiving a voice or speech signal regarding a request for a maintenance task transmitted from a first communication terminal inside the workroom using short-range wireless communication using a second communication terminal outside the workroom, and applying voice recognition to the received voice or speech signal,extracting a keyword from the request for a maintenance task acquired by speech recognition and performing a maintenance support suggestion to retrieve the maintenance task content based on the extracted keyword, and converting the result of retrieving the maintenance task content into a speech signal and transmitting the speech signal from the second communication terminal to the first communication terminal by short-distance wireless communication.

[0021] [Fig. 1] is a block diagram showing the maintenance management environment of a large number of paper sheet processing devices handling a large number of gold-exchangeable banknotes in a cash processing center of a bank as a paper sheet processing system according to this embodiment. In [Fig.l], a paper sheet processing system 11 is arranged inside a work room A managed under high security. The interior of the work room A and the exterior of the work room B are respectively provided with a first communication terminal 12 and a second communication terminal 13, both of which are of the mobile type and are configured to carry out short-distance wireless communication with each other by means of a speech signal.The work maintenance person M inside the work room A observes the maintenance information such as a knowledge number or the like to be displayed, for example, on a maintenance management display section of the paper sheet processing system 11, and reads aloud the content and description of the circumstance by putting the first communication terminal 12 carried by the maintenance person M into the transmission-enabled state to thereby transmit the information in the form of a speech signal.

[0022] On the other hand, upon receiving the speech signal transmitted from the first communication terminal 12 outside the workroom B, the second communication terminal 13 automatically informs a voice information processing device 14 of the speech signal. This voice information processing device 14 recognizes the speech of the speech signal sent from the second communication terminal 13, inputs maintenance information such as a knowledge number, description of the circumstance, and the like given by the maintenance person at work M, and informs a maintenance support system 16 of the inputted maintenance information and the time of receipt via a communication network 15.

[0023] The above-mentioned voice information processing device 14 is specifically constituted, as shown in [Fig.2], of a central processing unit (CPU) 141, a random access memory (RAM) 142, a read only memory (ROM) 143, a storage section 144, a communication interface section 145, a speech input / output interface section 146 all connected to a bus.

[0024] The CPU 141 reads a specified program from the ROM 143 and executes the program to thereby perform the specified processing. The RAM 142 provides the CPU 141 with a working space. The ROM 143 stores therein in advance the programs necessary for voice recognition, determination processing, and the like. The storage section 144 stores therein the information necessary for processing sheets of paper, processing results, and the like. The communication interface section 145 performs communication with the maintenance support system 16 and a maintenance term corpus 18 via the network 15. The speech input / output interface section 146 performs voice communication with the maintenance person at work M via the second communication terminal 13 and the first communication terminal 12.

[0025] The above-mentioned maintenance support system 16 is configured to manage the past maintenance results of the paper sheet processing system 11, and is configured to be capable of acquiring the past response example by inputting a keyword such as an error code, the constituent devices, the sensor, and the like, and performing the retrieval. Furthermore, the maintenance support system 16 has a function of evaluating the displayed content as a retrieval result after the maintenance task is completed. More specifically, the maintenance support system 16 has a function of performing a five-level evaluation as an evaluation method.

[0026] The above-mentioned maintenance support system 16 is connected to a knowledge database 17 in which past response examples are aggregated. This knowledge database 17 outputs the relevant data based on the retrieval function of the maintenance support system 16. In the past response example, information regarding the information (paper sheet type, processing capacity) finally processed by the paper sheet processing system 11 is recorded.

[0027] Furthermore, the maintenance support system 16 is connected to the maintenance term corpus 18 in which maintenance terms are aggregated via (or perhaps directly) the communication network 15. In this maintenance term corpus 18, technical terms (terms to be used very frequently during a maintenance operation such as "stoppage", "abnormal sound", "absence of movement", and the like) of the system 11 which is the object of maintenance are aggregated.

[0028] A flow of the maintenance task of a case in which the voice information processing device 14 and the maintenance support system 16 of the above-mentioned embodiment are used is shown in [Fig.3].

[0029] In [Fig. 3], upon detecting a voice communication between the second communication terminal 13 and the first communication terminal 12 (step S1), the voice information processing device 14 acquires the raw data of the voice communication of the maintenance person at work M, and detects a voice output (step S2). Here, the voice information processing device 14 extracts a keyword from the raw data, compares the extracted keyword with the maintenance term corpus 18 to extract a keyword necessary for retrieval (step S3), and executes retrieval using the keyword from the knowledge database 17 by means of the retrieval function of the maintenance support system (step S4).

[0030] A diagram of the maintenance support system and the state at the time of maintenance knowledge retrieval input is shown in [Fig.4]. That is, as shown in [Fig.4](a), a keyword consisting of the maintenance knowledge retrieval element is input, and retrieval is executed. Heretofore, this processing was performed by human work, but it is automated in this embodiment. When retrieval is completed, a list of the retrieval result is displayed as shown in [Fig.4](b), and the corresponding contents become selectable.

[0031] Specific examples of countermeasures for each knowledge number are shown in [Fig.5]. By selecting one of the knowledge numbers #1 to #6, and specifying the cause and phenomenon, it is possible to extract the elements such as the supposed cause, the processing content, the notation, the constituent devices, the evaluation, the type of paper sheet, the processing capacity and the like.

[0032] Then, the association degree and processing rate of the "keyword" are calculated from the retrieval result. It should be noted that the association degree and processing rate are defined as follows: # degree of association = number of keyword coincidences # processing rate = processing capacity # ideal processing capacity x 100.

[0033] Here, the processing rate indicates, as the value thereof decreases, that it took longer to complete the response maintenance action and that there is a high possibility that the example is a particular response example or a response example in which the amount of work to be performed by the maintenance person at work M is enormous. Furthermore, the processing capacity indicates the final amount processed by the paper sheet processing system 11 at the time of the past response example and the ideal processing capacity indicates the ideal amount that the paper sheet processing system 11 aims to process. These elements serve as indexes to be used when a response example is to be selected.

[0034] First, as a retrieval result, it is determined whether or not the corresponding data is present (step S5). When the corresponding data is present, the degree of association between the corresponding data and the keyword is calculated (step S6), the answer examples are rearranged in relation to the retrieval results in the order of the<degré d’association, évaluation> and, when the examples have the same<degré d’association, évaluation> , the examples are rearranged in descending order of processing rate (in descending order of influence to be exerted on the operation), whereby a list of information items each having a high possibility of being an example of a response at the time of occurrence of an emergency is acquired (step S7).Then, the acquired list of information items of the response examples is used to provide a "suggestion as to the gist of a device confirmation" to the work maintenance person M from the system, reduction of the response examples is started on the basis of the response content of the work maintenance person M, and the "suggestion as to the gist of a device confirmation" to be provided from the system to the maintenance person M and the response of the work maintenance person M are repeated, whereby it is finally possible for the work maintenance person M to acquire the target response example (step S8).

[0035] On the other hand, when the corresponding data is absent as a retrieval result, a keyword for the new entry is automatically entered, and the keyword is presented (step S9). Then, the confirmation of the content is executed by the maintenance person at work M, and the registration is performed (step S10), whereby the series of processing is completed.

[0036] More specifically, when the corresponding data are retrieved from the knowledge database 17 shown in [Fig.6A], the association degree and the processing rate are calculated and, if the retrieval result shown in [Fig.6B] is obtained, the association degrees (in descending order), the evaluations (in descending order), and the processing frequencies (in ascending order) are rearranged as shown in [Fig.6C], and information is provided to the maintenance person at work M from the acquired list of response examples.

[0037] Here, assuming that knowledge numbers #5 and #4 shown in [Fig.6D] are selected in order as an example, the system specifies the questions to be asked to the maintenance person at work M and the processing contents to be presented to the maintenance person M when acquiring the answers to the questions.

[0038] That is, in the state in which knowledge number #5 is selected, a question “Is the appearing phenomenon “occurrence of xx”?” is presented. At this time, when the maintenance person at work M responds with “Yes,” the system asks a question “Has a beam replacement been executed?” to the maintenance person M. At this time, when the maintenance person at work M responds with “No,” the system changes the answer example to be questioned due to the receipt of the answer “No” and asks a question “Has a confirmation of the operation of the conveyance device been executed?” to the maintenance person M based on the answer example of knowledge number #4, and when the maintenance person at work M responds with “Yes,” then the system asks a question “Has the controller replacement been executed?” to the maintenance person M.Here, when the maintenance person at work M responds with “Yes”, the system provides the guidance “It is conceivable that there is a need to “perform a confirmation of component AA and execute its replacement”. Please refer to the example response of “#4” to the maintenance person at work M.

[0039] By the flow described above, the retrieval is performed automatically based on the content of the conversation between the short-distance communication terminals without depending on the amount of knowledge about the maintenance operation and the degree of operating skill about the maintenance support system both possessed by the maintenance person on standby, whereby the human work outside the work room is eliminated and it is possible to reduce the operating cost. Furthermore, it is possible to acquire the retrieval result in the order of the<degré d’association, évaluation, et taux de traitement> , and to promptly provide the response content to be performed by the maintenance person at work to the maintenance person at work without the intervention of the waiting maintenance person.

[0040] As described above, according to this embodiment, the maintenance retrieval task to be performed outside the workroom is automated, and thus the operating efficiency is improved. Furthermore, a maintenance response having a high degree of association is automatically extracted, whereby operating efficiency is improved. In addition, the dependence on the operating skill of the maintenance person on standby regarding the maintenance support system is eliminated, and thus the operating efficiency is improved.

[0041] It should be noted that, in the above-mentioned embodiment, although the description has been given by taking the cash processing center of a bank as an example, the embodiment is also applicable to high-security cases where the introduction of communication devices other than mobile communication terminals for short-distance communication into the work room is prohibited, such as a nuclear power plant, a defense work site, and the like.

[0042] Furthermore, the invention configured to extract a keyword from the speech of the mobile communication terminal for short-distance communication is also applicable to other purposes. Examples include a telephone response from a remote location and the like. Furthermore, by changing the keyword extraction from speech to the keyword extraction from text, the above-mentioned invention can also be applied to the case in which a request is made by the maintenance person at work by means of a text medium such as a mail or the like from the maintenance person at work.

[0043] Although certain embodiments have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the inventions. Indeed, the original embodiments described herein may be embodied in various other forms; further, various omissions, substitutions, and modifications in the form of the embodiments described herein may be made without departing from the spirit of the inventions. The appended claims and their equivalents are intended to cover such forms or modifications as would fall within the scope and spirit of the inventions.

Claims

1. Claims A method for supporting a maintenance task of a paper sheet processing system (11) configured to support a task of a paper sheet processing system (11) used within a workroom of a cash processing center of a bank workroom, the method being implemented by a maintenance management system and characterized in that it comprises: receiving a voice or speech signal regarding a request for a maintenance task transmitted from a first communication terminal (12) inside the workroom using short-distance wireless communication using a second communication terminal (13) outside the workroom, and performing voice recognition of the received voice or speech signal; extracting a keyword from the request for a maintenance task acquired by speech recognition and executing a maintenance support suggestion to implement a step of retrieving maintenance task content based on the extracted keyword; and converting the result of retrieving the maintenance task content into a speech signal and transmitting the speech signal from the second communication terminal (13) to the first communication terminal (12) by short-distance wireless communication, said method further comprising: using a maintenance term corpus (18) in which maintenance terms associated with the maintenance task are aggregated to thereby recognize technical terms of a system that is an object of the maintenance task and using the technical terms for said retrieval of the maintenance task content based on the extracted keyword, said maintenance management system being further configured to implement the following steps: the entry of a new keyword, when it is determined as a result of said retrieval step that the corresponding data serving as an example response is absent, - a retrieval of content based on this keyword, notification of a result of said retrieval, confirmation of the content by a maintenance person and recording of the keyword.

2. A method for supporting a maintenance task of a paper sheet processing system (11) according to claim 1, characterized in that: the maintenance task content comprises a knowledge database (17) of past maintenance task support, the maintenance task content being retrieved by this database based on a retrieval function of the maintenance support system (16) and using a knowledge reference number as a keyword

3. A method for handling a maintenance task of a paper sheet processing system (11) according to claim 2, characterized in that it further comprises the following steps performed by the maintenance management system: - determining whether corresponding data serving as a response example is present or not as a result of said retrieving step; - calculating, when it is determined that the corresponding data is present, a degree of association indicating the number of coincidences between the corresponding data items and the keywords; - calculating a processing rate indicating a ratio between an actual processing capacity and a capacity corresponding to the processing amount targeted by the paper sheet processing system; - acquiring said response examples in descending order of the degree of association and the processing rate;- notifying a maintenance person of a suggestion relating to the maintenance task based on an acquired information list of response examples; and - reducing the response examples based on a result of a response to the suggestion.;

Images