JP6809765B2 - Document creation system related to services for work machines - Google Patents

Description

Regarding the document creation system related to services for work machines.

When a serviceman related to a work machine such as a crane car or a forklift has a trouble such as an accident or a breakdown in the work machine, he / she takes some measures to solve the event of the work machine caused by the trouble. Skilled service personnel are familiar with the treatment of various events in the work machine. On the other hand, inexperienced servicemen often do not know the appropriate treatment for this event, and it is difficult to easily recognize the appropriate treatment.

Therefore, for example, there is a failure diagnosis method as disclosed in Patent Document 1. The failure diagnosis unit of the machine tool control device disclosed in Patent Document 1 includes a unit storage unit, a phenomenon storage unit, a defect occurrence prediction component storage unit, a treatment method storage unit, a solution storage unit, and a failure diagnosis control. It has a department. The operator selects the unit in which the defect has occurred using the display unit consisting of the touch panel, selects the failure phenomenon corresponding to each unit, selects the component that can cause the defect corresponding to each phenomenon, and selects each component. Select the corresponding action method and narrow down the solution to the problem. As a result, even an inexperienced worker can respond to an event of a machine tool.

By the way, the serviceman needs to present a quotation regarding the trouble to the customer who is the owner of the work machine, and needs to submit a work report to the department to which the serviceman belongs. Therefore, the serviceman prepares these documents after dealing with the trouble of the work machine.

The task of creating these documents from scratch wastes a lot of effort and time for service personnel. Therefore, technologies such as a quotation preparation system and a quotation preparation method have been developed (see, for example, Patent Document 2). According to such a quotation issuing system, a serviceman can save labor and time as compared with creating a document from the beginning.

However, even in this case, the serviceman has to perform the document preparation work again after the work machine responds, which is still a waste of labor and time. Further, in the technique disclosed in the above patent document, there is a possibility that information is erroneously input and an erroneous document is created.

Japanese Unexamined Patent Publication No. 2017-68678 Japanese Unexamined Patent Publication No. 2003-208531

Therefore, the problem to be solved by the present invention is a document that can reduce the labor and time required for document creation and can create an accurate document by creating a document in conjunction with a service for troubles of a work machine. It is to provide a creation system.

In order to solve the above problems, the document creation system related to the service for the work machine of the present invention is
A storage unit that stores events that may occur in the work machine and actions corresponding to the events, and
A display unit that displays the entered information and
An input unit for inputting event information including the event generated in the work machine or a part of the event, and
An output unit that acquires a list of lower-level events or the actions corresponding to the input event information from the storage unit and outputs the information to the display unit.
An input / output storage unit that stores the input event information and the output treatment, and
A database containing work machine information about the work machine and document information about the document to be created,
A document creation unit that creates a document related to the event and the action based on the event that occurred in the work machine, the action corresponding to the event, the work machine information, and the document information.
A similarity calculation unit that calculates the similarity between the input event information and the event stored in the storage unit,
Unsupervised learning is performed using a neural network using a plurality of past events that have occurred in the work machine described in a plurality of reports as learning data, and when a word is input, the meaning of the input word. A trained neural network that outputs the distributed representation vector corresponding to
When the event information is input, the morphological analysis unit that divides the event information into words and
A word vectorization unit that generates a distributed representation vector for each word by the learned neural network from the event information divided for each word.
An event vectorization unit that generates a distributed representation vector corresponding to all the texts of the input event information from the distributed representation vector of each generated word.
A vector expression storage unit that stores a distributed expression vector generated in response to the event in the storage unit is provided.
The similarity calculation unit is similar based on the cosine similarity between the distributed expression vector generated in correspondence with the input event information and the distributed expression vector stored in the vector expression storage unit. Calculate the degree,
When the event information is selected and input from the event list, the output unit further acquires the event list or the action corresponding to the input event information from the storage unit, and the output unit obtains the event information from the storage unit. It is characterized in that it is output to the display unit, and a list of the events for which a predetermined degree of similarity or higher is calculated is output to the display unit .

The above document creation system is, for example,
The document is an estimate and / or report on the event occurring on the work machine and the action on the work machine.

The above document creation system is, for example,
Further including a terminal, a document creation device, and a server capable of communicating with the terminal and the document creation device.
The terminal has the display unit, the input unit, and the output unit.
The document creation device includes the database and the document creation unit.
The server has the storage unit, the output unit, and the input / output storage unit.

The document creation system is preferably
The distributed representation vector corresponding to the meaning of the input word is based on the distributed representation vector of the subword extracted by dividing the input word into N-grams.

The above document creation system is, for example,
The input unit further inputs basic information including any or all of serviceman information, date and time information, location information, and customer information regarding the serviceman M.
The input / output storage unit further stores the basic information input by the input unit.
The document creation unit further creates the document based on the basic information stored in the input / output storage unit.

The document creation system according to the present invention can reduce the time and labor required for document creation and create an accurate document by creating a document in conjunction with a failure diagnosis for a work machine.

It is an overall view which shows the schematic structure of one Embodiment of the document creation system which concerns on this invention. It is a block diagram of the document creation system of FIG. It is a figure which shows an example of the report which described the past event. It is a figure which shows an example of the input screen displayed on the display part of FIG. It is a figure which shows another example of an input screen. This is an example of the treatment displayed on the display unit of FIG.

Hereinafter, an embodiment of the document creation system S relating to the service for the work machine of the present invention will be described with reference to the drawings. The work machine in this embodiment is a forklift F, but it is merely an example and is not limited thereto.

FIG. 1 is an overall view showing a schematic configuration of a document creation system S according to the present embodiment. When a trouble such as an accident or failure occurs in the forklift F, the serviceman M identifies the cause of the trouble and the necessary measures by the troubleshooting terminal 2 capable of communicating with the server 1. The document creation device 3 acquires the input / output information of the terminal 2 from the server 1, and automatically creates a document D such as a report or a quotation based on this information.

<Server>
As shown in FIG. 2, the server 1 includes a storage unit 10, a morphological analysis unit 11, a word vectorization unit 12, a trained neural network (hereinafter, simply referred to as “learned NN”) 13, and an event vectorization. It has a unit 14, a vector expression storage unit 16, a similarity calculation unit 15, and an input / output storage unit 17. Further, the server 1 can communicate with the terminal 2 and the document creation device 3 with each other.

The storage unit 10 stores an event that may occur in the forklift F and a treatment corresponding to this event. The event and the action are extracted from a plurality of trouble cases including the event that occurred in the forklift F in the past and the action taken for the event. Events and actions may be extracted, for example, from the trouble cases described in previous report 300 (see FIG. 3).

When the event information is input by the input unit 21 (see the terminal 2 in FIG. 2), the morphological analysis unit 11 divides the input event information into words. The event information is a word group or a word that expresses an event that occurred in the forklift F as a sentence such as "the engine does not start" or a keyword such as "engine" or "does not start". The event information is divided by the morphological analysis unit 11 with a space for each word, and is divided and processed. For example, the event information that "the engine does not start" is divided into "engine", "ga", "starting", and "not". For example, "Mecab" may be used as a tool for morphological analysis. Further, a pre-made corpus for the forklift F may be used as a morphological analysis tool. This is preferable because each noun in the event information is appropriately extracted.

The word vectorization unit 12 converts the event information divided for each word by the morphological analysis unit 11 into a distributed expression vector for each word by the learned NN 13.

When a word is input, the learned NN13 outputs a distributed expression vector corresponding to the meaning of the input word. The learned NN13 is pre-learned without supervision using a neural network using the events described in a plurality of past reports 300 divided for each word as learning data, and the learned event information is divided for each word. When is input, a distributed expression vector corresponding to the meaning of each word is output based on the relationship between each word included in the event information.

The learned NN13 is when words with high relevance such as words that are likely to appear in the same event (co-occurrence) or words that are likely to appear in close positions in the same event are input. , Pre-learned to output a similar distributed representation vector. This is based on the distribution hypothesis (Harrs, 1954, Rubenstein & Goodenough, 1965) that words appearing in similar contexts are also semantically similar. As the word vectorization unit 12, for example, a language model such as “Word2Vec”, “Glove”, or “Fastext” based on the neural network technology may be used. For distributed representations and Word2Vec, see "MIKOLOV, Tomas, et al. Distributed representations of words and phrases and their compositionality. In: Advances in neural information processing systems. 2013. p. 3111-3119." For Grove, you can refer to "PENNINGTON, Jeffrey; SOCHER, Richard; MANNING, Christopher D. Glove: Global Vectors for Word Representation. In: EMNLP. 2014. p. 1532-1543." For Fastext, you can refer to "JOULIN, Armand, et al. Bag of tricks for efficient text classification. ArXiv preprint arXiv: 1607.01759, 2016."

As a method of outputting the distributed expression vector of each word, the learned NN13 preferably creates a distributed expression vector in consideration of the subwords included in each word. Specifically, the word vectorization unit 12 divides the input word into N-grams, extracts subwords, and uses the distributed expression vector corresponding to the meaning of the input word as the sum of the distributed expression vectors of the subwords or It is preferable to use an average. For example, in the case of the word "intake hose", the subword corresponds to "suction", "intake", "kiho", "ho", "su", and "su". As a result, the learned NN13 outputs a similar distributed expression vector to, for example, the “intake hose”, the “suction hose”, and the “hose”, so that the input event information can be changed to a notational fluctuation or an unknown word. Can also be handled. For example, "Fastext" may be used as a tool for creating a distributed representation vector in consideration of subwords.

As shown in FIG. 3, in the report 300, the event that occurred in the forklift F is described in the defect content column by each person in charge (serviceman M). Therefore, in the report 300, each event is described in various expressions in combination with the individuality of the person. By using the description of these events as learning data, the learned NN13 can appropriately convert each word included in the event information of various expressions into a distributed expression vector.

The event vectorization unit 14 is used by the word vectorization unit 12 for each word so that the similarity calculation unit 15 can calculate the similarity between the input event information and the event stored in the storage unit 10. Distributed representation Vectorized event information is converted into a distributed representation vector of event information units based on the distributed representation vector for each word. The event vectorization unit 14 may use the average of the distributed expression vectors of the words included in the event information as the feature vector of the event information itself as a method for converting the event information into a distributed expression vector. Alternatively, the event vectorization unit 14 may use feature vector averaging using SCDV (Sparse Composite Document Vectors) as a method for more accurately converting event information into a distributed representation vector. For SCDV, see "Dheeraj Mekala et al .: SCDV: Sparse Composite Document Vectors using soft clustering over distributional representations, Proc of EMNLP, 2017".

The similarity calculation unit 15 calculates the similarity between the input event information and the event stored in the storage unit 10. The events in the storage unit 10 are morphologically analyzed in advance by the morphological analysis unit 11, decentralized expression vectorization for each word by the word vectorization unit 12, and decentralized expression vectorization for each event by the event vectorization unit 14. There is. The distributed expression vectorized event is stored in the vector expression storage unit 16 in association with each event in the storage unit 10.

The similarity calculation unit 15 receives input event information based on the cosine similarity between the distributed expression vectorized event information and the distributed expression vectorized event stored in the vector expression storage unit 16. The degree of similarity with the event in the storage unit 10 is calculated. Cosine similarity is a similarity calculation method used when comparing documents in a vector space model. Since the cosine similarity expresses the closeness of the angles formed by the vectors, the input event information and the event in the storage unit 10 are more similar if the cosine similarity is close to 1, and should be close to 0. It will be less similar.

The input / output storage unit 17 stores information on input / output performed at the terminal 2.

<Terminal>
The terminal 2 has a display unit 20, an input unit 21, and an output unit 22. Although the terminal 2 is a tablet in the present embodiment, it is merely an example, and may be, for example, a notebook computer or a smartphone.

The display unit 20 displays the input information. The display unit 20 corresponds to a tablet display.

The input unit 21 is a tool for inputting event information and the like. In the case of a tablet, the display is a touch panel, which also serves as a display unit 20 and an input unit 21. The input unit 21 may be, for example, a mouse or a keyboard. The input unit 21 inputs basic information and event information including any or all of serviceman information, date and time information, location information, and customer information regarding the serviceman M. Further, an event (event information) is selected and input by the input unit 21 from the event list displayed on the display unit 20 by the output unit 22.

The output unit 22 acquires a list of lower-level events or actions corresponding to the input event information from the storage unit 10 and outputs the list to the display unit 20. In addition, the output unit 22 also outputs to the display unit 20 a list of events for which a degree of similarity of a predetermined value or higher is calculated by the degree of similarity calculation unit 15. Further, when an event is selected and input from the event list, the output unit 22 acquires a list of events or actions corresponding to the input event from the storage unit 10 and outputs the list to the display unit 20.

<Document creation device>
The document creation device 3 has a database 30, a document creation unit 31, and a document output unit 32. The document creation device 3 includes, for example, a personal computer. The document creation device 3 may be provided, for example, in the room of the department to which the serviceman M belongs.

The database 30 is stored in the storage unit of the document creation device 3. The database 30 includes forklift information regarding the forklift F and document information regarding the document D to be created.

The forklift information includes information such as the brand, type, model, model, and part of the forklift F. The forklift information also includes the amount of service related to each forklift F for preparing a quotation. Services for the forklift F include, for example, analyzing an event that has occurred in the forklift F and taking action on the event.

The document information includes information such as a template necessary for creating the document D.

The document preparation unit 31 prepares an estimate and / or a report regarding these events and treatments based on the events occurring in the forklift F, the treatments for the forklift F, the forklift information, and the document information. The event that occurred in the forklift F and the action for the forklift F are included in the input / output information of the terminal 2 stored in the input / output storage unit 17 of the server 1. Therefore, the document creation unit 31 acquires the information regarding the event that occurred in the forklift F and the action for the forklift F by receiving the input / output information from the server 1.

In the case of a quotation, the document preparation unit 31 describes the events that occurred in the forklift F and the measures for the forklift F in the template of the quotation included in the document information, and also describes the events and the amount of money corresponding to the measures. To do. The input / output storage unit 17 of the server 1 also includes basic information including any or all of serviceman information, date / time information, and location information regarding the serviceman M input by the input unit 21 of the terminal 2. The document creation unit 31 receives these basic information from the server 1 and describes it in the template of the quotation. As a result, a quotation is created.

The document output unit 32 is a monitor, a printer, or the like, and displays or prints the document D created by the document creation unit 31.

<Operation of document creation system>
The configuration of the document creation system S according to the present embodiment has been described above. Next, the specific operation of the document creation system S according to the present embodiment will be described with reference to the drawings.

When an event due to a trouble occurs in the forklift F, the serviceman M inputs basic information or the like to the terminal 2 by the input unit 21. The information input by the input unit 21 is transmitted to the server 1. The input / output storage unit 17 stores the received input information.

As shown in FIG. 4, when the screen 400 for inputting the forklift information and the event information is displayed on the display unit 20, the serviceman M displays the brand, type, model, and model of the forklift F in the image areas 400a to 400d. , The part where the event occurred, the error code of the event is directly input or selectively input.

Next, as shown in FIG. 5, a screen 500 for selecting a top event is displayed on the display unit 20, and when the event information 500a and 500b are displayed, one of the event information is selected by the serviceman M. Entered. At this time, the event information corresponding to the event generated in the forklift F may not be included in the displayed event information. In this case, the serviceman M inputs the event information to the keyword input area 400e shown in FIG.

The similarity calculation unit 15 calculates the similarity between the input event information and the event stored in the storage unit 10, and transmits the event for which the similarity is calculated to be equal to or higher than a predetermined value to the terminal 2. The output unit 22 outputs a list of received events to the display unit 20 and displays it. Next, when the serviceman M selects and inputs an event corresponding to the event generated in the forklift F from the displayed event list, the output unit 22 stores the list of events corresponding to the input event in the storage unit 10. Is received from and output to the display unit 20.

As shown in FIG. 6, when the serviceman M selects and inputs the lowest event, the output unit 22 outputs and displays the treatment 600a corresponding to this event on the display unit 20. Even an inexperienced serviceman M can identify a treatment corresponding to an event occurring in the forklift F by referring to this treatment. The action output by the output unit 22 is transmitted to the server 1 and stored in the input / output storage unit 17.

The document creation unit 31 receives the basic information stored in the input / output storage unit 17, the event that occurred in the forklift F, and the action for the forklift F. Next, the document creation unit 31 describes the received information and the forklift information corresponding to the received information in the template of the document D for creating the received information. As a result, document D is created.

According to the document creation system according to the present embodiment, the serviceman M does not need to create the document D again after the forklift F responds. Moreover, since the document D is created from the input / output information of the terminal 2, there is no erroneous input of information, and the possibility that an erroneous document is created is reduced.

Although one embodiment of the document creation system S according to the present invention has been described above, the present invention is not limited to the above embodiment.

(1) In the document creation system S, the terminal 2 may have the configuration of the server 1. In this case, the document creation device 3 receives basic information, an event that has occurred in the work machine, and a measure for the work machine from the input / output storage unit 17 of the terminal 2.

(2) The storage unit 10 may further store technical data corresponding to the treatment. In this case, the output unit 22 may output the technical data corresponding to this treatment to the display unit 20 together with the treatment and display it on the display unit 20. As a result, even an inexperienced service person can take appropriate measures.

1 Server 10 Storage unit 11 Morphological analysis unit 12 Word vectorization unit 13 Learned neural network 14 Event vectorization unit 15 Similarity calculation unit 16 Vector expression storage unit 17 Input / output storage unit 2 Terminal 20 Display unit 21 Input unit 22 Output unit 3 Document creation device 30 Database 31 Document creation unit 32 Document output unit 300 Report

Claims (5)

A document creation system related to services for work machines
A storage unit that stores events that may occur in the work machine and actions corresponding to the events, and
A display unit that displays the entered information and
An input unit for inputting event information including the event generated in the work machine or a part of the event, and
An output unit that acquires a list of lower-level events or the actions corresponding to the input event information from the storage unit and outputs the information to the display unit.
An input / output storage unit that stores the input event information and the output treatment, and
A database containing work machine information about the work machine and document information about the document to be created,
A document creation unit that creates a document related to the event and the action based on the event that occurred in the work machine, the action corresponding to the event, the work machine information, and the document information.
A similarity calculation unit that calculates the similarity between the input event information and the event stored in the storage unit,
Unsupervised learning is performed using a neural network using a plurality of past events that have occurred in the work machine described in a plurality of reports as learning data, and when a word is input, the meaning of the input word. A trained neural network that outputs the distributed representation vector corresponding to
When the event information is input, the morphological analysis unit that divides the event information into words and
A word vectorization unit that generates a distributed representation vector for each word by the learned neural network from the event information divided for each word.
An event vectorization unit that generates a distributed representation vector corresponding to all the texts of the input event information from the distributed representation vector of each generated word.
A vector expression storage unit that stores a distributed expression vector generated in response to the event in the storage unit is provided.
The similarity calculation unit is similar based on the cosine similarity between the distributed expression vector generated in correspondence with the input event information and the distributed expression vector stored in the vector expression storage unit. Calculate the degree,
When the event information is selected and input from the event list, the output unit further acquires the event list or the action corresponding to the input event information from the storage unit, and the output unit obtains the event information from the storage unit. A document creation system characterized in that it outputs to the display unit and further outputs a list of the events for which a predetermined degree of similarity or more is calculated to the display unit . The document creation system according to claim 1, wherein the document is an estimate and / or a report relating to the event occurring in the work machine and the action on the work machine. Further including a terminal, a document creation device, and a server capable of communicating with the terminal and the document creation device.
The terminal has the display unit, the input unit, and the output unit.
The document creation device includes the database and the document creation unit.
The server has the storage unit, the output unit, and the input / output storage unit.
The document creation system according to claim 1 or 2. Any of claims 1 to 3, wherein the distributed expression vector corresponding to the meaning of the input word is based on the distributed expression vector of the subword extracted by dividing the input word into N-grams. The document creation system described in item 1. By the input section, further serviceman information about the service Ma emissions, date and time information, place information, the basic information including any or all of the customer information is entered,
The input / output storage unit further stores the basic information input by the input unit.
The document creation according to any one of claims 1 to 4, wherein the document creation unit further creates the document based on the basic information stored in the input / output storage unit. system.

Images