JP5432941B2 - Emergency response work support apparatus and emergency response work support method - Google Patents

Description

  The present invention relates to an emergency countermeasure business support apparatus and an emergency countermeasure business support method, and is particularly capable of handling an unexpected event that is not described in a manual in an emergency such as a large-scale disaster. The present invention relates to an emergency countermeasure business support apparatus and an emergency countermeasure business support method.

  The person in charge of disaster countermeasures in companies and local governments must give appropriate instructions to each department of the organization in the event of an emergency such as a disaster and flexibly respond to the crisis they face. Further, in the event of an emergency, the person in charge of each department needs to take a different action from the normal work, and must execute the unsteady work in parallel. In preparation for such a situation, it is important to prepare a crisis response manual in which response actions are made into procedures. 2. Description of the Related Art Conventionally, a system has been disclosed in which a crisis response manual is digitized and mounted on a computer, and the execution of response actions described therein is supported (see, for example, Patent Document 1). There is also a technique disclosed as a command support system for a person in charge of a decision making position (see, for example, Patent Document 2).

  However, although the number of occurrences of a disaster case itself may be small, it is rare to occur in exactly the same situation without reproducibility. Also, unexpected situations often occur, so there is a possibility that a situation not in the manual will occur. Therefore, in actual disaster response operations, it is necessary to take countermeasure actions in an order different from the manual, or it becomes necessary to re-plan the operation, and the disaster countermeasure person in charge depends on the situation at that time. The action to be taken will be reconsidered based on experience and general social common sense. In this way, there is also an activity to leave a rare case of actual correspondence that is not in the manual to the future generations (for example, see Non-Patent Document 1). The recording method here is a format in which the contents of the memories at that time are spoken in time series as they are, and it is not a configuration or mechanism that can be referred to in actual emergency response operations.

  On the other hand, there is disclosed a technique for accumulating business correspondence cases and tips as business know-how that cannot be specified in a manual (see, for example, Patent Document 3). In these techniques, in order to refer to the stored text, a keyword is set for the stored searched text, and the search keyword input in the keyword search and the keyword set for the searched text are determined. When only matching items are extracted, or when texts are registered, it is necessary to build a relationship between the work in the manual and the work know-how related to the work in advance. However, in keyword search, only texts with matching keywords can be used to refer to texts describing past know-how. In other words, even if the text has a similar context, it may not be extracted unless the keyword words match. In reality, the operations described in the manual are regularly reviewed and added and deleted repeatedly. For this reason, in the method of constructing the relationship, it is necessary to rebuild the relationship between the manual and the business know-how each time.

  A concept search technique is known as a search technique that does not rely on keywords (see, for example, Patent Document 4). However, in the conventional concept search, no specific examination has been made to associate the manual with the business know-how in an emergency.

JP 2009-265686 A JP 2010-204712 A Japanese Patent No. 2853206 Japanese Patent No. 4499003

Haruo Hayashi, 2 others, "Decision Factors for Disaster Prevention Disaster Ethnography", 1st edition, Japan Broadcasting Corporation (NHK Publishing), December 20, 2009, p.1-13

  When dealing with a rare and non-reproducible event called disaster response, past response cases and business know-how are important information. However, in reality, such information is accumulated only as personal experience and is not in a form that can be referred to by a third party. Furthermore, local governments have prepared plans and manuals in advance for the occurrence of disasters, but for the reasons described above, it is difficult to verify and verify their effectiveness. However, once a disaster occurs, the person in charge is forced to respond. In other words, it is required to create a database of past experiences, to refer to past experiences according to the situation in the event of a disaster, and to extract necessary tasks from existing manuals according to the situation. If such a thing becomes possible, it will be possible to respond continuously to events that are not described in the manual in a rare and non-reproducible disaster.

  In view of such problems, the object of the present invention was the current situation for the user when an unexpected event not described in the manual occurred in an emergency such as a large-scale disaster. To provide an emergency response work support apparatus and an emergency response work support method capable of presenting past information.

  An emergency countermeasure business support apparatus according to the present invention includes a business flow storage unit that stores business units in which emergency response processing procedures are documented, a business flow indicating a transition relationship between the business units, and past emergency situations. A business know-how storage unit that stores business know-how in which correspondence examples of the business are documented; a concept vector generation unit that generates a concept vector of the business unit and the business know-how; and a log storage unit that stores a reference history of the business know-how , Based on the concept vector and the reference history, a business know-how extracting unit that extracts a first business know-how related to the first business unit, and a business know-how selected by the user from the first business know-how A business unit search unit for searching for a second business unit.

  The emergency countermeasure business support device according to the present invention further includes a business know-how search unit for searching for a second business know-how related to the first business know-how, wherein the business know-how search unit includes the first business know-how search unit. Search for business know-how related to the first business know-how using the business know-how as search input text, and then repeatedly search for business know-how related to the searched business know-how using the searched business know-how as search input text. The searched business know-how is used as the second business know-how, and the business unit search unit searches for the business unit related to the business know-how selected by the user from the first business know-how and the second business know-how. The second business unit is preferably used.

  The emergency response work support apparatus according to the present invention further includes a work flow formulation unit that creates a new work flow in which a business unit selected by the user from the second business unit is added to the current work flow. It is desirable to provide.

  The log storage unit may increase the number of times the business know-how is included in the reference history when the user refers to the business know-how for a predetermined time or more.

  In addition, the business know-how search unit uses the searched business know-how as a search input text, and when performing the iterative processing to search for the business know-how related to the searched business know-how, the business know-how previously obtained as the search input text Is searched as the related business know-how, it is desirable to end the repetitive processing.

  As described above, the solution of the present invention has been described as an apparatus. However, the present invention can be realized as a method, a program, and a storage medium storing the program, which are substantially equivalent thereto, and the scope of the present invention. It should be understood that these are also included.

  For example, the emergency response business support method according to the present invention is an emergency response business support method by an emergency response business support device that supports a user by presenting information in an emergency, and a processing procedure by the emergency response business support device Is a step of storing business units documented emergency processing procedures, a business flow showing the transition relationship between the business units, and storing business know-how documenting past emergency response cases A step of generating a concept vector of the business unit and the business know-how, a log storing step of storing a reference history of the business know-how, and the first business unit based on the concept vector and the reference history The first business know-how to be extracted, and the business know-how selected by the user from the first business know-how A search step of searching a second operational units, is intended to include.

  The processing procedure by the emergency countermeasure business support apparatus further includes a search step for searching for a second business know-how related to the first business know-how, wherein the first business know-how is obtained in the search step. Search for business know-how related to the first business know-how as a search input sentence, and then repeatedly search for business know-how related to the searched business know-how using the searched business know-how as a search input text. Each business know-how is set as the second business know-how, and in the search step, a business unit related to the business know-how selected by the user from the first business know-how and the second business know-how is searched and the second business know-how is searched. It is desirable to be a business unit.

  The processing procedure by the emergency countermeasure business support apparatus may further include a step of formulating a new business flow in which the business unit selected by the user from the second business unit is added to the business flow being executed. desirable.

  In the log storing step, when the user refers to the business know-how for a predetermined time or more, it is desirable to increase the number of times the business know-how is included in the reference history.

  In the search step, when the iterative processing for searching for business know-how related to the searched business know-how is performed using the searched business know-how as the search input text, the business know-how that has previously become the search input text is When searching for related business know-how, it is desirable to end the repetitive processing.

  According to the emergency response work support apparatus and the emergency response work support method according to the present invention, in the event of an unexpected event that is not described in the manual in an emergency situation such as a large-scale disaster, It is possible to present past information that matches the current situation.

FIG. 1 is a diagram showing a schematic configuration of a system including an emergency countermeasure work support apparatus according to an embodiment of the present invention. FIG. 2 is a diagram showing a schematic configuration of an emergency countermeasure work support apparatus according to an embodiment of the present invention. FIG. 3 is a diagram illustrating an example of a business flow and a business unit. FIG. 4 is a diagram illustrating an example of business know-how. FIG. 5 is a diagram illustrating an example of a business know-how reference history. FIG. 6 is a diagram illustrating an example of an index conscious of the use of business know-how. FIG. 7 is a diagram illustrating an example of business flow formulation. FIG. 8 is an operation flowchart of the emergency countermeasure work support apparatus shown in FIG.

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

  FIG. 1 is a diagram showing a schematic configuration of an emergency countermeasure business support system including an emergency countermeasure business support device according to an embodiment of the present invention. The emergency response work support device 1 of the present invention supports a user by presenting information in an emergency. The emergency countermeasure work support apparatus 1 is connected to a user client terminal 2 (2a to 2c) via a network. The emergency countermeasure business support apparatus 1 may display a business flow on the screen of the client terminal 2 through the network, or acquire an operation (for example, a business unit selection operation) performed by the user on the client terminal 2. it can. The user of the client terminal 2 performs the emergency countermeasure work along the work flow displayed on the screen. In addition, when there is no business unit corresponding to the current situation, the user operates the client terminal 2 to instruct the emergency countermeasure business support apparatus 1 to reconstruct the business flow, and related information (business unit Business know-how).

  FIG. 2 is a diagram illustrating a schematic configuration of the emergency countermeasure work support apparatus 1. The emergency countermeasure business support apparatus 1 includes a storage unit 10 and a control unit 20, and the storage unit 10 includes a business flow storage unit 11, a business know-how storage unit 12, a concept vector storage unit 13, and a log storage unit. 14, the control unit 20 includes a business flow execution unit 21, a concept vector generation unit 22, a business know-how extraction unit 23, a business know-how search unit 24, a business unit search unit 25, and a business flow formulation unit 26. With.

  The business flow storage unit 11 stores a business flow that expresses a standard procedure for emergency response as a unit for each unit and expresses a transition relationship. Hereinafter, a unit that is included in the business flow and that documents the emergency response procedure is referred to as a business unit. The business flow indicates a transition relationship between business units.

  FIG. 3 is a diagram illustrating an example of a business flow and a business unit. The business flow includes business units U01 to U07. The rectangular business units (U01, U02, U04 to U07) indicate processing items performed by the user, and the diamond-shaped business units (U03) indicate user judgment items. Further, specific work items such as U071 corresponding to U07 are recorded in each of the units U01 to U07, and the user can execute a corresponding work on-site based on the work items.

  The business know-how accumulating unit 12 stores business know-how in which past emergency response cases (time series information, personal experience, related knowledge, etc.) are documented. FIG. 4 is a diagram illustrating an example of business know-how. For example, when there is no business unit corresponding to the current situation in the business flow, the user operates the client terminal 2 to transmit a business flow reconstruction instruction to the emergency countermeasure business support device 1. Thereby, the business know-how related to the business unit being executed in the emergency countermeasure business support apparatus 1 is searched, and the user can acquire the business know-how related to the current situation.

  The concept vector storage unit 13 stores the concept vector generated by the concept vector generation unit 22. The details of the concept vector will be described later, but since it is possible to determine that sentences with close concept vectors are highly related, by using the concept vector, business know-how related to the currently executing business unit and other Business units can be searched.

  The log storage unit 14 stores a reference history of business know-how. That is, the log storage unit 14 stores the number of times the user has referred to each business know-how as an index of the effectiveness of the business know-how. FIG. 5 is a diagram illustrating an example of a reference history (log information) stored in the log storage unit 14. FIG. 5A is log information indicating the total number of searches for business know-how, and records the date and time of the search and the cumulative number of times each time the search for business know-how is executed. The business know-how search includes both business know-how extraction by the business know-how extraction unit 23 and business know-how search by the business know-how search unit 24. FIG. 5B is log information indicating the number of times business know-how is referenced, and records the date and time of registration of business know-how, the number of characters, and the number of times of reference for each business know-how text ID. Here, the reference count indicates the number of times the user actually selected the business know-how when presented to the user as business know-how that is highly likely to be related to the current situation. Note that simply displaying business know-how on the screen does not make it possible to ascertain whether or not the user has actually read the contents, so if the user references the business know-how for a certain period of time (the business know-how is displayed on the screen for a certain period of time or more). The number of references may be increased only in the case of That is, for example, a standard silent reading speed such as 0.1 second per character is set in advance, and a standard sentence reference time obtained by multiplying the number of characters in the log information of FIG. 5B is obtained. The reference count may be incremented by one. In this case, if the standard speed is set to 0, the reference count increases by 1 when the reference screen is opened.

  Further, the log storage unit 14 can store an index that is conscious of the use of business know-how. FIG. 6 is a diagram illustrating an example of an index conscious of the use of business know-how. After referring to the selected business know-how, the user can register the user's evaluation for the items shown in FIG. The evaluation points registered by the user can be used as an index of the effectiveness of business know-how, such as the number of times of reference to each business know-how described above.

  The business flow execution unit 21 reads the business flow from the business flow storage unit 11 and presents the business flow to the client terminal 2 via the network. The client terminal 2 displays a business flow on the screen, and performs processing such as highlighting the business unit being executed in accordance with the progress of the user's work, for example.

  The concept vector generation unit 22 generates each business unit in the business flow stored in the business flow storage unit 11 and the business know-how concept vector stored in the business know-how storage unit 12. Here, the concept vector is an index indicating the degree of association between sentences used for concept search. The concept vector generation unit 22 analyzes business unit and business know-how texts and arranges them as points in a multidimensional concept vector space. The similarity in the context in the concept vector space is expressed by the fact that the frequency distribution of words appearing in the sentence is similar, that is, the directions of the vectors are close. The concept vector generation unit 22 stores the created concept vector in the concept vector storage unit 13. The concept vector generation unit 22 can automatically create a business unit and business know-how concept vector in a predetermined cycle or the like, or can create a concept vector in real time when executing a search. For example, even when the business unit and business know-how text are partially changed, the concept vector generation unit 22 can automatically calculate the latest concept vector. Concept search and concept vectors are techniques known to those skilled in the art, and in this application, detailed explanations such as theoretical explanations, specific analysis methods, and relevance levels will not be given. See Reference 4.

  The business know-how extracting unit 23 extracts the business know-how stored in the business know-how storage unit 12 based on the concept vector stored in the concept vector storage unit 13 and the business know-how reference history stored in the log storage unit 14. To do. Specifically, the business know-how extracting unit 23 is executing an execution based on the “business relevance” that is an index obtained by multiplying the relevance representing the closeness of the concept vector and the reference rate calculated from the reference history. Extract business know-how related to business units.

First, _{assuming that} the concept vector of the search input sentence e _{k for} the concept search is v (e _k ) and the concept vector of the searched sentence d _{i to} be searched is v (d _i ), the business know-how extracting unit 23 calculates the formula 1 makes it possible to obtain the degree of association R (k, i) between concept vectors. The value of Expression 1 is a value from -1 to 1, and the closer to 1, the more similar the search input sentence and the searched sentence are.

The business know-how extracting unit 23 further obtains the business know-how reference history from the log storage unit 14, and multiplies the relevance R (k, i) by the reference rate obtained from the reference history as shown in Equation 2, The business relevance P (k, i), which is an index that adds usefulness to the past reference history, is calculated. The business know-how extracting unit 23 extracts business know-how related to the business unit being executed using the business relevance level P (k, i). Here, n (d _i) is the number of times obtained by subtracting the cumulative number of searches of the registration date and time earlier from the cumulative number of times the search text d _i of the total number of searches of the registration date and time later, that is at the moment of the search text d _i, The influence of the number of searches before the text is registered is removed. l (d _i ) is the number of times the searched text d _i is referenced. That is, l (d _i ) / n (d _i ) represents the reference rate of the searched sentence d _{i with} respect to the total number of searches, and the higher the reference rate of the searched sentence d _i , the larger the value of the index of Expression 2 is. Become.

  The business know-how extracting unit 23 causes the client terminal 2 to display a list of business know-how extracted in descending order of the business relevance P (k, i) of Equation 2. The business know-how extracting unit 23 further normalizes and multiplies the right side of Equation 2 by the average value of the evaluation points as shown in FIG. 6 (in the case of the 6-step evaluation shown in FIG. 6, divides by 3). You can also. Thereby, an index conscious of the use of business know-how can be reflected in the extraction result. Since the user may want to narrow down the search by looking at the extracted result, the business know-how extracting unit 23 has a keyword addition function, and adds the sentence added by the user to the search input sentence. The business know-how extraction process can be executed repeatedly. The business know-how extracting unit 23 may extract the business know-how by a conventional keyword search instead of the concept search.

  After the user selects the business know-how extracted by the business know-how extracting unit 23, the business know-how search unit 24 searches the business know-how related to the business know-how selected by the user as a search input sentence. Is repeatedly searched for related business know-how. In other words, the business know-how search unit 24 repeatedly executes the business know-how search using the business know-how selected by the user as the base and sequentially searches the business input with the highest degree of business relevance, and the business know-how group targeted for the search input text To the user.

  Specifically, the business know-how search unit 24 searches for business know-how based on the business relevance shown in Equation 2, similarly to the business know-how extraction unit 23. The business know-how search unit 24 newly sets a text of business know-how having the maximum business relevance among the search results as a search input text, and repeats the same search process. Even if the business know-how searching unit 24 is set to repeat the search process 100 times, for example, when the business know-how that has previously become a search input sentence has been searched as related business know-how, The search repetitive process can be terminated with. It should be noted that whether or not the search process by the business know-how searching unit 24 is executed after the business know-how extracting unit 23 extracts the business know-how can be set in advance. The automatic search processing by the business know-how search unit 24 can efficiently extract related sentence groups from a large number of sentence groups, and the system automatically collects related sentences without setting the threshold of the distance of the feature space. It becomes possible to determine and extract the condition. The business know-how search unit 24 may search for business know-how by a conventional keyword search instead of a concept search.

  The business unit search unit 25 executes a concept search using the business know-how selected by the user among the business know-how searched by the business know-how extraction unit 23 or the business know-how search unit 24 as a search input sentence. A list of business units having a high degree of relevance is presented to the user. The business unit search unit 25 may search for business units by a conventional keyword search instead of a concept search.

  The business flow formulation unit 26 incorporates the business flow group selected by the user into the current business flow among the business flows including the business unit selected by the user from the business unit list presented by the business unit search unit 25. It has. That is, the business flow formulation unit 26 can formulate a new business flow in which the business unit selected by the user from the business unit list presented by the business unit search unit 25 is added to the business flow being executed. The business flow formulation unit 26 stores the created new business flow in the business flow storage unit 11.

  FIG. 7 is a diagram illustrating an example of business flow formulation by the business flow formulation unit 26. As shown in FIG. 7A, the business unit search unit 25 can present the entire business flows A, B, and C including the searched business units A ′, B ′, and C ′ as a list to the user. The user selects business units A ′ and C ′ considered to be necessary according to the current situation from the list. In this case, not only the business units A ′ and C ′ extracted by the search by the business unit search unit 25 but also other business units in the business flow can be selected. As shown in FIG. 7B, the business flow formulation unit 26 formulates a new business flow E by incorporating the selected business units A ′ and C ′ into the business flow D currently being executed. At this time, the business flow formulation unit 26 may formulate a new business flow E by adding business units A ′ and C ′ in the order selected by the user based on the business unit D ′ currently being executed. it can. The business flow formulation unit 26 stores the new business flow E in the business flow storage unit 11. In addition, the business flow formulation unit 26 transmits information on the new business flow E to the business flow execution unit 21, and the business flow execution unit 21 determines the business flow D currently being executed and the newly created business flow E. To be presented to the user. The user returns to the workflow execution screen again and executes the newly created corresponding task immediately before.

  FIG. 8 is an operation flowchart of the emergency countermeasure work support apparatus 1. The business know-how and business unit concept vectors used in the concept search are generated in advance by the concept vector generation unit 22 or generated in real time when the search is executed.

  First, the business flow execution unit 21 selects and reads a business flow that matches the current crisis event from the business flow storage unit 11 (step S101), and displays the business flow and the business unit on the screen of the client terminal 2 via the network. Is displayed (step S102). The business flow execution unit 21 determines whether or not there is a business flow restructuring instruction from the user of the client terminal 2 (step S103), and the corresponding business progresses according to the displayed business flow (that is, the business flow) When there is no restructuring instruction), when the user has executed all business units in the business flow, the process is terminated (step S104).

  However, if an event that does not correspond to the currently displayed business flow occurs during execution of a business unit, the user determines that it is necessary to reconstruct the future corresponding business. In this case, the business flow execution unit 21 can confirm the business flow restructuring instruction by using, for example, a control message notified from the client terminal 2 (Yes in step S103). In restructuring the business flow, the control unit 20 determines what to search for among the business units and business know-how (step S105). This determination can be made with reference to, for example, the initial setting of the emergency response work support apparatus 1 or a control message notified from the client terminal 2. Here, the control unit 20 determines whether to directly search for business units, to first search for business know-how, to search for business units from the reference business know-how, or to execute both in parallel. It can be performed.

  In step S105, when the business unit is a search target, the business unit search unit 25 stores the text of the business unit currently being executed by the user as a search input text used for the concept search, and is stored in the business flow storage unit 11. A business unit included in the business flow is searched (step S106). For example, when the user is executing the business unit U02 of the business flow shown in FIG. 3, the business unit search unit 25 searches and inputs the sentence “The information group collects and organizes information and reports to the headquarters”. Perform a concept search as a sentence. The business unit search unit 25 presents the search result to the client terminal 2 via the network, and the search result is displayed on the screen of the client terminal 2 (step S107). Here, when a related business unit is not searched, or when the user performs a search further narrowing down the current situation, the business unit search unit 25 searches for a search condition (keyword or keyword) newly input by the user. The business unit is re-searched by adding (text) (step S109). For example, when the user wants to specify what information is targeted, the business unit search unit 25 adds “situation of road closure due to earth and sand disaster” and the like to the search condition. Further, the business unit search unit 25 may execute a keyword search using a keyword directly input by the user instead of the concept search.

  On the other hand, when the business know-how is to be searched in step S105, the business know-how extraction unit 23 stores the text of the business unit currently being executed by the user in the business know-how storage unit 12 as a search input text used for the concept search. The obtained business know-how is extracted (step S110). Here, the business know-how extracting unit 23 extracts business know-how based on the “business relevance” described above. The business know-how extracting unit 23 presents the extraction result to the client terminal 2 via the network, and the extraction result is displayed on the screen of the client terminal 2 (step S111). Here, when the related business know-how is not extracted, the business know-how extracting unit 23 re-extracts the business know-how by adding a search condition (keyword or sentence) newly inputted by the user (step) S113).

  When the extraction result displayed on the screen of the client 2 has business know-how related to the business unit currently being executed, the user selects the business know-how. The control unit 20 determines the business know-how (selected business know-how) selected by the user (step S114), and determines whether to further search for business know-how related to the selected business know-how (step S115). When performing business know-how search, the business know-how search unit 24 sequentially searches for business know-how using the selected business know-how as a search input sentence and using the business relevance level (step S116). The business know-how searching unit 24 presents the search result to the client terminal 2 through the network, and the search result is displayed on the screen of the client terminal 2 (step S117).

  When the extraction result or search result displayed on the screen of the client 2 has business know-how related to the business unit currently being executed, the user selects the business know-how. The control unit 20 determines the business know-how (selected business know-how) selected by the user (step S118), and the business unit search unit 25 performs a concept search for business units composed of sentences similar to the selected business know-how. Execute (step S119). The business unit search unit 25 presents the search result to the client terminal 2 through the network, and the search result is displayed on the screen of the client terminal 2 (step S120).

  The user selects a business unit considered necessary from the business units displayed on the screen of the client 2. The business flow formulation unit 26 determines the business unit selected by the user (step S121), adds the business unit selected by the user to the business flow currently being executed, and formulates a new business flow (step S122). The business flow formulation unit 26 stores a new business flow in the business flow storage unit 11. Further, the business flow formulation unit 26 transmits information on a new business flow to the business flow execution unit 21, and the business flow execution unit 21 reads the newly formulated business flow and returns to the loop processing of the business unit.

  Note that the process of extracting / searching business know-how text by executing the concept search in step S110 and step S116 can be replaced by a conventional keyword search when the user determines to directly execute a general keyword search. .

  As described above, according to the present embodiment, the business know-how extracting unit 23 extracts business know-how related to the business unit being executed based on the concept vector and the reference history, and the business unit search unit 25 extracts the business know-how. The business unit related to the business know-how selected by the user from the business know-how is searched. As a result, in an emergency such as a large-scale disaster, when an unexpected event that is not described in the manual (business flow and business unit) occurs, past information on the current situation for the user (Related business know-how / business unit) can be presented.

  In the past, it was necessary to link business know-how with related business units. However, according to this embodiment, when registering business know-how, it is not necessary to link with existing business units, and it is simply registered as text. Just do it. The registered business know-how is automatically linked with other business units and business know-how using concept search technology. Further, conventionally, when searching for business know-how in the current situation, only texts that match the input keyword can be extracted by keyword search, but according to this embodiment, conventional keyword search is performed. In addition, using conceptual search technology, the most relevant text as a whole of the business know-how and business unit group registered at that time (appears in similar contexts, that is, search input text and searched text) Searching for a similar pattern of appearance frequency between words to be performed).

  Moreover, according to this embodiment, even if the text of the business unit is corrected, the relevance with the registered business know-how can be automatically updated each time. In other words, even if there are many unexpected operations such as disaster response and manual maintenance is insufficient, if past cases and know-how are accumulated as business know-how text, they are automatically associated with the manual. It can be utilized. In addition, although it is not known whether the text extracted simply by concept search or keyword search is information useful for business, according to the present embodiment, a business manual with high evaluation based on the number of times the user has referred to in the past ( A sentence frequently referred to by a user). In this way, users can extract useful sentences efficiently from a large group of sentences by combining not only indicators that sentences are similar but also indicators that are useful sentences. Become.

  Further, according to the present embodiment, the business know-how search unit 24 searches for business know-how related to the business know-how selected by the user as a search input text, and thereafter relates to the searched business know-how as a search input text. The business unit search unit 25 repeatedly searches for business know-how, and the business unit search unit 25 searches for business units related to the business know-how selected by the user from the business know-how extracted by the business know-how extraction unit 23 and the business know-how searched by the business know-how search unit 24. To do. This makes it possible to sequentially search and present information that can be related to the current situation to the user.

  Further, according to the present embodiment, the business flow formulation unit 26 formulates a new business flow in which the business unit selected by the user from the business units searched by the business unit search unit 25 is added to the business flow being executed. . As a result, the user can quickly create a new manual based on the referenced information, and can proceed with a plan for a specific response work.

  In addition, when the user refers to the business know-how for a certain time or more, the log storage unit 11 increases the number of times the business know-how is included in the reference history. Thereby, since the reference rate of the business know-how actually referred to by the user increases, it becomes possible to present more useful business know-how to the user.

  Further, the business know-how searching unit 24 uses the searched business know-how as a search input sentence, and when performing the iterative process for searching for related business know-how, if the business know-how that has previously become the search input sentence is searched, The process ends. Thereby, it is possible to prevent redundant repeated processing and efficiently present related business know-how to the user.

  Although the present invention has been described based on the drawings and examples, it should be noted that those skilled in the art can easily make various modifications and corrections based on the present disclosure. Therefore, it should be noted that these variations and modifications are included in the scope of the present invention. For example, the functions included in each component, each step, etc. can be rearranged so that there is no logical contradiction, and multiple components, steps, etc. can be combined or divided into one It is.

DESCRIPTION OF SYMBOLS 1 Emergency response work support device 10 Storage part 11 Business flow storage part 12 Business know-how storage part 13 Concept vector storage part 14 Log storage part 20 Control part 21 Business flow execution part 22 Concept vector generation part 23 Business know-how extraction part 24 Business know-how Search unit 25 Business unit search unit 26 Business flow formulation unit 2a, 2b, 2c Client terminal

Claims (10)

A business flow storage unit that stores business units that document emergency response processing procedures, and a business flow that indicates a transition relationship between the business units;
A business know-how storage unit that stores business know-how that has been documented in the past cases of emergency response;
A concept vector generation unit that generates a concept vector of the business unit and the business know-how;
A log storage unit for storing a reference history of the business know-how;
A business know-how extracting unit that extracts a first business know-how related to the first business unit based on the concept vector and the reference history;
A business unit search unit for searching for a second business unit related to the business know-how selected by the user from the first business know-how;
Emergency response work support device comprising: A business know-how search unit for searching for a second business know-how related to the first business know-how;
The business know-how search unit searches for business know-how related to the first business know-how using the first business know-how as a search input text, and thereafter searches the business know-how as a search input text. It repeatedly searches for business know-how related to know-how, and each searched business know-how becomes the second business know-how,
2. The business unit search unit according to claim 1, wherein the business unit search unit searches for a business unit related to the business know-how selected by the user from the first business know-how and the second business know-how and sets it as the second business unit. Emergency response work support device.   The emergency countermeasure work according to claim 1 or 2, further comprising a business flow formulation unit that formulates a new business flow in which a business unit selected by the user from the second business unit is added to the business flow being executed. Support device.   The emergency according to any one of claims 1 to 3, wherein the log accumulation unit increases the number of times the business know-how is included in the reference history when the user refers to the business know-how for a certain period of time or longer. Countermeasure work support device.   The business know-how searching unit uses the searched business know-how as a search input text, and when performing the iterative processing to search for the business know-how related to the searched business know-how, the business know-how previously converted into the search input text The emergency countermeasure work support apparatus according to any one of claims 2 to 4, wherein when it is searched as related work know-how, the repetitive processing is terminated. An emergency response work support method by an emergency response work support device that supports a user by presenting information in an emergency,
The processing procedure by the emergency response work support device is as follows:
Storing a business unit that documents the emergency response processing procedure and a business flow indicating a transition relationship between the business units;
Storing business know-how that documented past emergency response cases;
Generating a concept vector of the business unit and the business know-how;
A log storage step for storing a reference history of the business know-how;
Extracting a first business know-how related to a first business unit based on the concept vector and the reference history;
A search step for searching for a second business unit related to the business know-how selected by the user from the first business know-how;
Emergency response work support method including The processing procedure by the emergency countermeasure business support device further includes a search step for searching for a second business know-how related to the first business know-how,
In the searching step, the business know-how related to the first business know-how is searched using the first business know-how as a search input text, and thereafter, the searched business know-how is used as the search input text. It repeatedly searches for related business know-how, and each searched business know-how becomes the second business know-how,
The emergency according to claim 6, wherein, in the search step, a business unit related to the business know-how selected by the user from the first business know-how and the second business know-how is searched for to be the second business unit. Time support work support method.   The processing procedure by the emergency countermeasure business support apparatus further includes a step of formulating a new business flow in which a business unit selected by the user from the second business unit is added to the business flow being executed. Or the emergency countermeasure work support method described in 7.   The emergency according to any one of claims 6 to 8, wherein, in the log storage step, when the user refers to the business know-how for a certain time or more, the number of times of reference to the business know-how included in the reference history is increased. Countermeasure work support method. In the search step, when performing the iterative process of searching for the business know-how related to the searched business know-how using the searched business know-how as the search input text, the business know-how previously used as the search input text is related. The emergency countermeasure work support method according to any one of claims 7 to 9, wherein when the search is made as work know-how, the repetitive processing is terminated.

Images