JP6502816B2 - Planning support system and planning support method - Google Patents

Description

  The present invention relates to a plan support system and a plan support method, and more specifically to a technology that enables accurate and efficient planning based on the planning reason of a planner.

  In the case of manufacturers, etc., there is a task of drawing up a production plan of a product based on its own business resources. As a technique related to such planning, for example, production lots serving as allocation candidates are evaluated based on an empirical rule database constructed by selection results of past allocation candidates for each evaluation item of production condition information of the production lots. The first evaluation value and the second evaluation value which is the prediction result of the production result obtained by performing the production planning simulation on each production line in order from the one with the highest first evaluation value to the production lot that becomes the allocation candidate Based on the third evaluation value, which is the evaluation result of the allocation candidate production lot and the allocation candidate production line, which are obtained from Assign a specific production lot among the candidates to the production line, and learn based on the specific production lot And production planning method for updating the base (see Patent Document 1) has been proposed.

JP, 2012-194712, A

  However, when a planner makes a plan, it may be based on knowledge (hereinafter referred to as a plan reason) based on various information as well as its own business resources. In that case, a plan different from the plan that was created by considering only business resources will be drawn up. That is, the contents of the plan also change depending on the planner's reason which varies depending on the planner and the situation at that time.

  However, in the prior art, it is not possible to make a plan by reflecting such a planning reason on an empirical rule. In addition, there is no need to present the reason for planning automatically created plans. In such a case, other planners may not only understand the planning reason of the corresponding plan, but may not know the important planning reason or misidentify and make an inappropriate plan correction.

  Therefore, an object of the present invention is to provide a technology that enables accurate and efficient planning based on the planning reason of the planner.

The planning support system of the present invention for solving the above problems corresponds to the planning reason of the planned plan and the planning algorithm configured including the learning model constructed using the machine learning technology used for the plan. The matching process is performed by collating the storage device added and stored, the reference information acquired from the predetermined address on the predetermined network, and the planning reason of each plan, and information matching the reference information identified in the matching process For the planning reason included in the above, the planning algorithm associated with the planning reason is specified, and for the planning algorithm, predetermined information on the next planned event is input to acquire the planning result, and And an arithmetic device for outputting the planned result to a predetermined device.

Further, in the plan support method of the present invention, a plan creation algorithm configured by including a planning reason of a planned plan and a learning model constructed using a machine learning technique used for the plan is stored in association with each other. An information processing system provided with a storage device collates reference information acquired from a predetermined address on a predetermined network with the reason for creating each plan, performs matching processing, and matches the reference information identified in the matching processing. respect drafting reasons contained information is to identify the planning algorithm associated with the drafting reasons, with respect to the planning algorithm, and obtains the planning results by entering predetermined information on Imatsugi plans events Outputting the planned result to a predetermined device.

  According to the present invention, it is possible to create a highly accurate and efficient plan based on the planning reason of the planner.

It is a figure showing composition of a plan support system in this embodiment. It is a figure showing composition of an automatic planning system in this embodiment. It is a figure showing the example of hardware constitutions of the plan support system in this embodiment. It is a figure which shows the structure of order information DB in this embodiment. It is a figure which shows the structure of internal information DB in this embodiment. It is a figure which shows the structure of temporary plan DB in this embodiment. It is a figure which shows the structure of execution plan DB in this embodiment. It is a figure which shows the structure of correction reason DB in this embodiment. It is a figure which shows the example 1 of a flow of the plan assistance method in this embodiment. It is a figure which shows the example 1 of a screen in this embodiment. It is a figure which shows the example 2 of a screen in this embodiment. It is a figure which shows Screen example 3 in this embodiment. It is a figure which shows the example 4 of a screen in this embodiment. It is a figure which shows Screen example 5 in this embodiment. It is a figure which shows the example 2 of a flow of the plan assistance method in this embodiment. It is a figure which shows the example 3 of a flow of the plan assistance method in this embodiment.

--- System configuration ---

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a diagram showing a configuration of a plan support system 104 of the present embodiment. The plan support system 104 illustrated in FIG. 1 is a computer system that enables highly accurate and efficient planning based on the planning reason of the planner.

  The plan support system 104 in the present embodiment is a server device operated by a manufacturer or the like, and can be assumed to be communicably connected to an external network such as the Internet 101. In addition, this plan support system 104 exchanges data with the operator terminal 110 operated by a planner who plans a product manufacturing plan, and outputs processing information such as a provisional plan, an actual plan, a plan reason, etc. It becomes a thing.

  In addition, the plan support system 104 can use the order information DB 102 and the internal information DB 103 as a database conventionally managed by the manufacturer. Each of these databases 102 and 103 may be held in the storage device of the plan support system 104, or may be held by an existing system such as a production management system of a manufacturer and provided accessible to the plan support system 104. It may be

Among them, the order information DB 102 is a database storing order information of a predetermined product received from a customer of the manufacturer. The order information held by the order information DB 102 includes information such as the type, quantity, and delivery date of the product ordered from the customer. Details of the order information DB 102 will be described later.

  In the present embodiment, the database is configured by a general PC (Personal Computer) and general DB management software, and the DB management software provides an information search and update function and the like.

  Further, the internal information DB 103 is a database storing personal identification information such as identification information and years of service of workers, etc. who are in charge of the manufacturing process by the manufacturer, and sensor data representing the operation status of the production equipment. Details of the internal information DB 103 will be described later.

  On the other hand, the plan support system 104 of this embodiment includes the external information input unit 105, the order information input unit 106, the internal information input unit 107, the learning device selection unit 108, the automatic planning system 109, and the temporary plan correction unit 111 as functional units. , And an execution plan output unit 114. Further, the plan support system 104 holds a temporary plan DB 112, an execution plan DB 113, and a correction reason DB 115 as a database provided in its own storage device.

  The above-described units 105 to 109, 111, and 114 are functional units that are implemented by the CPU executing the program held by the plan support system 104.

  Of the functional units included in the plan support system 104, the external information input unit 105 acquires, from a predetermined information processing apparatus on the Internet 101, information serving as a reference for planning, that is, external information serving as reference information. Means to provide Here, the external information corresponds to information such as weather information and raw material price which can be obtained by requesting a predetermined server on the Internet 101.

  The order information input unit 106 is means for acquiring order information from the order information DB 102 described above and providing the order information to the automatic planning system 109. Further, the internal information input unit 107 is means for acquiring internal information from the above-described internal information DB 103 and providing the same to the automatic planning system 109.

  Further, the learning device selection unit 108 is means for receiving selection of a learning device by a planner and providing information of the learning device to the automatic planning system 109. Here, the learning device is a planning algorithm including a learning model constructed using a machine learning technology such as a neural network. As a general operation, when known (or similar to known) input data is input to a learning device which is this planning algorithm, the known (or similar to known) input It will output a known (or similar to known) output conventionally obtained from the data, that is, a planned result.

  In addition, the automatic planning system 109 is a system that creates a temporary plan based on the information provided from the external information input unit 105, the order information input unit 106, and the internal information input unit 107, and stores this in the temporary plan DB 112. is there. The automatic planning system 109 also has a predetermined learning function using data obtained from an execution plan DB 113 and a correction reason DB 115, which will be described later. The learning function will be described later. The details of the execution plan DB 113 and the correction reason DB 115 will be described later.

  The temporary plan correction unit 111 receives the correction operation of the temporary plan by the planner from the terminal for worker 110, and stores the temporary plan corrected by the operation in the execution plan DB 113 as an execution plan (the same applies hereinafter). Means to

  The execution plan output unit 114 is means for outputting the execution plan read out from the execution plan DB 113 to a predetermined device (for example, a management device of a production line) of a production line of a product.

  Next, the configuration of the automatic planning system 109 included in the plan support system 104 will be described in detail. FIG. 2 is a diagram showing the configuration of the automatic planning system 109 in the present embodiment.

  The automatic planning system 109 in the present embodiment is an apparatus for automatically creating a temporary plan, and includes a temporary plan creating apparatus 201 and a learning device 205.

  Among them, the temporary plan creation device 201 includes an external / internal information processing unit 202, a learning device adoption unit 203, and a temporary plan creation unit 204. The external / internal information processing unit 202 is means for performing matching processing between external information and internal information and the conventional planning reason (held by the correction reason DB 115). Further, the learning device adopting unit 203 is means for selecting the above-mentioned learning device according to the result of the above-mentioned matching process. Also, the temporary plan creation unit 204 is a means for creating a temporary plan using the learning device adopted by the learning device adoption unit 203.

On the other hand, the learning device 205 is a means for causing the learning device to learn the relation between the reason for correction, ie, the correction reason and the execution plan (output), which is corrected by the planner regarding the temporary plan. The learning device 205 includes a correction reason processing unit 206, an execution plan processing unit 207, and a learning unit 208. Among them, the correction reason processing unit 206 is means for reading the correction reason from the correction reason DB 115, matching with the conventional planning reason, and selecting a learning device linked to the matched conventional planning reason. Further, the execution plan processing unit 207 is means for reading the execution plan from the execution plan DB 113 and inputting it to the learning device. Also, the learning unit 208 is means for constructing a new learning device and adjusting the accuracy of the learning device.
--- Hardware configuration ---

  FIG. 3 is a diagram showing an example of the hardware configuration of the plan support system 104 in the present embodiment. The hardware configuration of the plan support system 104 in this embodiment is as follows. The plan support system 104 is held in a storage device 308 configured of an appropriate non-volatile storage element such as a solid state drive (SSD) or a hard disk drive, a memory 302 configured of a volatile storage element such as a RAM, and the storage device 308 The program 310 is read out to the memory 302 or the like to execute overall control of the apparatus itself and to perform various determinations, calculations and control processing as well as a CPU 301 (arithmetic apparatus) At least an interface 304 is provided. Each of these devices is connected by an interface 303 ("I / F" in the figure), which is an internal bus or the like.

  In addition, when the plan support system 104 adopts a configuration that directly receives an operation from a planner, the plan support system 104 further includes a keyboard 305 and a mouse 307 for receiving an input operation from the planner, and a display 306 for displaying processing data.

  Among them, the mouse 307 moves the pointer on the screen displayed on the display 306, and causes the planner to press a button at an arbitrary place to designate the position on the screen of the display 306 and transmit some action to the CPU 301. It is an apparatus. The mouse 307 can be replaced by a general touch panel. In this case, the normal pointer is not necessary.

Note that functional units (external information input unit 105 to learning unit selection unit 108, temporary plan correction unit 111, execution plan output unit 114) necessary as the plan support system 104 of this embodiment are mounted in the storage device 308. In addition to the program 310 and the OS 311 (Operating System) providing the execution environment of the program 310, at least each of the above-described databases (provisional plan DB 112, execution plan DB 113, correction reason DB 115) is stored.

  Also, the program 310 described above is configured to be introduced to the storage device 308 to the memory 302 by a removable storage medium or by the network interface 304 via a communication medium (communication line or carrier wave on the communication line). You may.

When a storage medium is adopted, a configuration in which an apparatus for reading the storage medium is connected to the interface 303 is used. As such a storage medium and a device for reading it, an optical disc (CD, DVD, Blu-ray disc, etc.) and its reader, a portable storage device using a flash memory, etc. can be assumed.
--- Data structure example ---

  Next, an example of a data structure in a database used by the plan support system 104 of the present embodiment will be described. FIG. 4 is a view showing the data structure of the order information DB 102 in the present embodiment.

  The data structure in the order information DB 102 is a collection of records in which the order delivery date 402, the quantity 403, and the category 404 information of the order are associated using the order number 401 as a key. Among these values, the value of the order number 401 is an identification number that uniquely identifies an order. Further, the value of the delivery date 402 is a value indicating the deadline for delivering the product for the corresponding order. For example, it represents the number of days until shipment. Further, the quantity 403 is a value representing the required number of products to be ordered. The type 404 is identification information that uniquely identifies the type of product designated by the customer in the corresponding order. Each item described above may include other data related to the order as data necessary for creating a plan.

  FIG. 5 is a view showing the data structure of the internal information DB 103 in the present embodiment. The internal information DB 103 includes a manufacturer information DB 501 and operation information of a manufacturing machine that manufactures the corresponding product in response to the above-mentioned order. Among them, the manufacturer information DB 501 is a set of records in which the names 504 of the workers actually in charge of the work on the manufacturing line and the values of the years of service 505 of the workers are associated.

  Further, the operation information of the manufacturing machine is a collection of records in which the operation information of the corresponding machine obtained from the sensor provided on the manufacturing line is associated with each manufacturing machine. Specifically, it is configured to include, as the operation information of the machine, observation values obtained from the sensors of the sensors X1 and Y1 and the sensors X2 and Y2 of the machine X1 and Y1 with respect to the manufacturing machine # 1 502. The operation information generated by each sensor is uploaded in real time from the management apparatus or the like of the manufacturing line to the plan support system 104, and stored in the internal information DB 103.

  In addition to the manufacturer information DB 501 and the operation information of the manufacturing machine as described above, the internal information DB 103 may include information such as the raw materials used for manufacturing and the stock status of products.

  FIG. 6 is a diagram showing the data structure of the temporary plan DB 112. As shown in FIG. The provisional plan DB 112 is a database in which the provisional plan generated by the automatic planning system 109 is stored. The provisional plan is a plan for producing a product corresponding to the order indicated by the order information DB 102 by a predetermined quantity and a predetermined delivery date, and has not been confirmed or corrected by the planner.

The data structure is a collection of records in which the due date 402, the quantity 403, the type 405, and the device number 602 are associated using the temporary manufacturing order 601 and the order number 401 as keys.

  Among them, the provisional manufacturing order 601 is an identification number uniquely identifying the manufacturing order of the orders in the provisional plan. Further, each value of the order number 401, the delivery date 402, the quantity 403, and the type 405 is the same value as that indicated by the corresponding record of the order information DB 102. The device number 602 is an identification number of a manufacturing machine at the time of manufacturing the corresponding order on the manufacturing line, and corresponds to the machine numbers of “1st machine” and “2nd machine” indicated by the corresponding record of the internal information DB 103.

  FIG. 7 shows the data structure of the execution plan DB 113. As shown in FIG. The execution plan DB 113 is a database in which the planner makes appropriate corrections and fixes the above-mentioned temporary plan, that is, a database for storing the execution plan. The data structure is a collection of records in which the due date 402, the quantity 403, the type 405, and the device number 702 are associated using the manufacturing order 701 and the order number 401 as keys.

  Among these, the production order 701 is an identification number that uniquely identifies the production order of orders in the execution plan. Further, each value of the order number 401, the delivery date 402, the quantity 403, and the type 405, and the device number 702 is the same value as that indicated by the corresponding record of the temporary plan DB 112.

  FIG. 8 shows the data structure of the correction reason DB 115. As shown in FIG. The revision reason DB 115 includes the reason for drafting each plan (whether the provisional plan or the execution plan) (the drafter has revised the reason for drafting the provisional plan or the reason for drafting the execution plan already determined). It is a database in which each information of the external information which is reference information at that time, the acquisition destination of the internal information, and the learning device used to generate the plan are stored in association with each other.

  The data structure is associated with the reason for correction of plan planning reason (including the concept of planning reason without correction), external information link destination 802, internal information link destination 803, and learning device number 804 It is a collection of records.

Among these, the correction reason 801 is described as an arbitrary character string. Also, the external information link destination 802 indicates the address of the web site published by the web server on the Internet 101. Further, an internal information link destination 803 indicates a storage destination address of internal information in the internal information DB 103. The learning device number 804 is an identification number that uniquely identifies the learning device associated with the correction reason 801.
--- Flow example 1--

  Hereinafter, the actual procedure of the plan support method in the present embodiment will be described based on the drawings. Various operations corresponding to the plan support method described below are realized by the program 310 which the plan support system 104 reads out to the memory 302 and executes. The program 310 is composed of code for performing various operations described below.

  FIG. 9 is a diagram showing a flow example 1 of the plan support method in the present embodiment. The flow is, for example, a flow that starts when a predetermined instruction from a planner is received from the operator terminal 110 or when storage of new order information in the order information DB 102 is detected. . This detection is performed by the plan support system 104 accessing the order information DB 102 at regular intervals, monitoring the presence or absence of addition and updating of records.

In the situation where the flow is started, the CPU 301 of the plan support system 104 instructs the external information input unit 105, the order information input unit 106, and the internal information input unit 107 to start data reading (1401). Communication between the CPU 301 and each part is generally
It can be executed by a PC (Remote Procedure Call) or the like, and in the following, it is assumed that communication between devices is performed by such a method.

  The external information input unit 105 that has received the data reading instruction refers to the value of the external information link destination 802 of each record in the correction reason DB 115, and updates the latest information such as weather information from a predetermined device on the Internet 101 indicated by the value. It will acquire external information.

  Similarly, the order information input unit 106 that has received the data reading instruction specifies the order information for a predetermined period (for example, one week) stored most recently in the order information DB 102 with reference to the time stamp or the like, I will get it.

  Also, the internal information input unit 107 that has received the data reading instruction refers to the value of the internal information link destination 803 of each record in the correction reason DB 115, and detects each sensor data etc. from a predetermined area of the internal information DB 103 indicated by the value. You will get the latest internal information.

  In step 1401, the CPU 301 of the planning support system 104 stores the external information, the order information, and the internal information acquired by the external information input unit 105, the order information input unit 106, and the internal information input unit 107 in the memory 302. Read

  When the external information input unit 105 acquires external information, the CPU 301 of the planning support system 104 stores the URL (external information) held by each record of the correction reason DB 115 in a general HTML (Hyper Text Markup Language) source acquisition tool. The HTML source is downloaded from a storage area of a predetermined device (a web server on the Internet 101) indicated by the URL given a value indicated by the link destination 802).

  The CPU 301 of the planning support system 104 which has obtained this HTML source from the above-mentioned HTML source acquisition tool executes a character string search based on predetermined keywords such as "typhoon", "heavy rain" and "tornado" in the HTML source, Character strings such as "Typhoon comes in 2 days" can be specified, and external information such as typhoon arrival information can be acquired.

  In addition, based on technology such as DOM (Document Object Model), as the value of the external information link destination 802, a form may be adopted in which the planner designates the specific part of the HTML source via the worker terminal 110 . In this case, information as to which tag information of the HTML source is read as external information is further added as a value of the external information link destination 802.

  Subsequently, the CPU 301 of the plan support system 104 executes creation of a provisional plan by providing the automatic planning system 109 with the external information, the order information, and the internal information read into the memory 302 in step 1401 described above ( 1402). Details of this processing in the automatic planning system 109 will be described later using the flow of FIG.

  Next, the temporary plan correction unit 111 of the plan support system 104 generates screen data including information on the temporary plan created by the automatic planning system 109, and outputs the screen data to the worker terminal 110 for display (1403). For example, it is possible to assume that the screen data is generated by setting various information related to the provisional plan in advance in accordance with the rules of the screen template with respect to the screen template held in the storage device 308 in advance.

On the other hand, the worker terminal 110 receives screen data including information on a provisional plan from the plan support system 104, displays it on a display, and provides it for the planner to view. It is assumed that the screen displayed on the display includes a display of each piece of information involved in the preparation of a plan, such as a provisional plan and its planning reason, order information, external information, and internal information.

  An example of the screen 900 thus displayed on the display of the worker terminal 110 is shown in FIG. This screen 900 is an input screen on which the temporary plan correction unit 111 of the plan support system 104 receives the correction operation of the planner on the temporary plan in the worker terminal 110.

  The screen 900 shown in FIG. 10 is a planner number input box 901, a temporary plan display box 902, a temporary plan correction box 903, a change button 904, a delete button 905, a temporary plan reason display box 906, a learner selection button 907, and execution. Plan display box 908, conventional correction reason selection button 909, correction reason input box 910, search input box 911, search button 912, external / external link destination input box 913, internal information display box 914, internal information link destination input box 915, And an object such as a save button 916.

  Among these, the planner number input box 901 is an input object for receiving personal number input such as an employee number from the planner. A temporary plan display box 902 is a display object for displaying the temporary plan created by the automatic planning system 109.

  Further, a temporary plan correction box 903 is an object for receiving a correction to the temporary plan displayed in the temporary plan display box 902.

  A replacement button 904 in the temporary plan correction box 903 is a button for replacing the corresponding order in the temporary plan display box 902 at the position corresponding to the button 904 with another order. In addition, the screen which changes when the planner presses (clicks) this replacement button 904 will be described later. The delete button 905 in the temporary plan correction box 903 is a button for deleting the corresponding order in the temporary plan display box 902 located at the position corresponding to the button 904.

  In addition, a temporary planning planning reason display box 906 is a display object for displaying the planning reason of the temporary planning. Further, a learning device selection button 907 is a button for selecting a learning device. In addition, the screen which changes when the planner has pressed the learning device selection button 907 will be described later.

  Also, the execution plan display box 908 is a display object for displaying the provisional plan corrected by the planner as an execution plan. The conventional correction reason selection button 909 is a button for displaying the conventional correction reason (planning reason). In addition, the screen which changes when the planner has pressed the conventional correction reason selection button 909 will be described later.

  Further, the correction reason input box 910 is an input object for receiving an input of a correction reason (a correction of the conventional planning reason, of course it may be a completely new planning reason) from the planner.

  The search input box 911 is an input object for receiving an input from the planner regarding a search keyword of external information. The search button 912 is a button for the planner to instruct the search of the external information. If the search button 912 is pressed, the CPU 301 of the plan support system 104 performs a keyword search by a general search engine on the Internet 101. In addition, the screen which changes when the planner presses the search button 912 will be described later.

Also, the external information link destination input box 913 is an input object for receiving an input regarding the external information link destination from the planner. An internal information display box 914 is a display object for displaying internal information. In addition, an internal information link destination input box 915 is an input object for receiving an input regarding the internal information link destination from the planner. Further, a save button 916 is a button for receiving an instruction of a data save operation by the planner.

  Here, the description returns to the flow of FIG. In step 1403 described above, the temporary plan correction unit 111 of the plan support system 104 displays the screen 900 on the worker terminal 110 as described above, and urges the planner to confirm the temporary plan and the planning reason (1403).

  For example, it is assumed that the planner recognizes that the content browsed on the screen 900 needs to be corrected. And suppose that the point which should be corrected was a plan reason. Then, the planner presses the learning device selection button 907 on the screen 900 to display the screen 1000 (FIG. 11) on the worker terminal 110. At this time, in response to the pressing of the learning device selection button 907, the worker terminal 110 transmits a request of the screen 1000 to the plan support system 104, and acquires and displays the screen data of the screen 1000.

  FIG. 11 shows a screen 1000 output in response to the pressing of the above-described learning device selection button 907. The screen 1000 includes objects of a learning device display box 1101, a selection button 1102, and a temporary plan creation button 1103.

  Among them, the learning device display box 1101 is a display object for displaying the type of learning device. Further, the selection button 1102 is a button for receiving from the planner the selection operation for the learning device displayed in the above-described learning device display box 1101. The temporary plan creation button 1103 is a button for instructing the automatic planning system 109 to perform a plan creation operation using the learning device selected by the pressing operation of the selection button 1102 described above. In response to the pressing of the button 1103, the automatic planning system 109 of the planning support system 104 will create a temporary plan using the learning device.

  When the above-described selection button 1102 is pressed by the planner via the worker terminal 110, the temporary plan correction unit 111 of the plan support system 104 identifies that the corresponding learning device is selected, and the screen 900 is displayed. With respect to the temporary plan and the planning reason to be shown, processing of correlating the information of the learning device with the record of the corresponding planning reason (planning reason displayed in the temporary planning planning reason display box 906) of the correction reason DB 115 is executed (1404).

  In addition, when the above-described temporary plan creation button 1103 is pressed by the planner via the worker terminal 110, the temporary plan correction unit 111 of the plan support system 104 instructs the temporary plan creation using the corresponding learning device. The system is given to the automatic planning system 109, the temporary plan is regenerated, and this is displayed in the temporary plan display box 902 (1405).

  When the planner recognizes that the temporary plan thus generated needs to be corrected, the planner operates the terminal 110 for worker to temporarily display the temporary plan displayed in the temporary plan display box 902. The pressing operation on the replacement button 904 and the delete button 905 of the plan correction box 903 is executed, and the screen 1100 (FIG. 12) is displayed on the operator terminal 110.

  At this time, in response to the pressing of the change button 904 or the delete button 905, the worker terminal 110 transmits a request of the screen 1100 to the plan support system 104, and acquires and displays the screen data of the screen 1100.

FIG. 12 shows a screen 11000 output in response to the pressing of the above-mentioned replacement button 904 or delete button 905. The screen 1100 includes an order information display box 1001 for displaying order information included in the temporary plan, and a selection button 1002.

  The order information display box 1001 is a display object for displaying order information other than the order in the temporary plan display box 902. Further, the selection button 1002 is a button for replacing the order information in the temporary plan display box 902 with the order in the order information display box 1001.

  When the planner presses the selection button 1002 corresponding to a certain order, the temporary plan correction unit 111 of the planning support system that has received the order has already received the pressing of the change button 904 on the screen 900, and the selection button A process of replacing the order in the order information display box 1001 in which the button 1002 is pressed with the temporary plan display box 902 on the screen 900 is executed.

  Further, it is assumed that the planner executes the pressing operation on the delete button 905 of the temporary plan correction box 903 with respect to the temporary plan displayed in the temporary plan display box 902. At this time, in response to the pressing of the delete button 905, the worker terminal 110 transmits, to the planning support system, a deletion request for the corresponding order from the tentative planning.

  The temporary plan correction unit 111 of the plan support system 104 that has received this executes processing to delete the information of the corresponding order to be deleted from the temporary plans displayed in the temporary plan display box 902.

  Thus, the temporary plan correction unit 111 of the plan support system 104 executes a process of replacing or deleting the temporary plan according to the intention of the planner (1406).

  In addition, when the planner recognizes that the correction is necessary for the reason for planning the temporary plan displayed in the temporary plan planning reason display box 906 of the screen 900, the planer operates the worker terminal 110. Then, the pressing operation for the conventional correction reason selection button 909 is executed, and the screen 1200 (FIG. 13) is displayed on the operator terminal 110.

  At this time, in response to the pressing of the conventional correction reason selection button 909, the worker terminal 110 transmits a request of the screen 1200 to the plan support system 104, and acquires and displays the screen data of the screen 1200.

  FIG. 13 shows a screen 1200 output in response to the pressing of the conventional correction reason selection button 909 described above. The screen 1200 includes a conventional correction reason display box 1201 and a selection button 1202.

  Among these, the conventional correction reason display box 1201 is a display object for displaying the existing planning reason which the correction reason DB 115 holds. The selection button 1202 is a button for a planner to select the planning reason displayed in the conventional correction reason display box 1201. When the planner presses the selection button 1202, the temporary plan correction unit 111 of the plan support system 104 that receives this via the operator terminal 110 recognizes it as a selection instruction of the corresponding planning reason, and the planning reason Is displayed in the correction reason input box 910.

  In this case, the temporary plan correction unit 111 of the plan support system 104 sets the planning reason displayed in the above correction reason input box 910 as the correction reason, and associates it with the learning device stored in the correction reason DB 115 in step 1404. Instead of the planning reason, it is registered in the corresponding record (1407).

  Subsequently, the temporary plan correction unit 111 of the plan support system 104 registers the temporary plan corrected in the above steps as an execution plan in the execution plan DB 113 (1408).

  In addition, the temporary plan correction unit 111 of the plan support system 104 transmits, to the worker terminal 110, an input request for each link destination of the external information and the internal information that is the basis of the correction reason described above, The input value of each link destination of the external information and the internal information by the person is acquired from the worker terminal 110 (1409).

  In this case, the above-mentioned planner operates the worker terminal 110 to input a keyword for searching for external information in the search input box 911 of the screen 900 and presses the search button 912. The temporary plan correction unit 111 of the plan support system 104 that has received this pressing via the worker terminal 110 returns the screen 1300 (FIG. 14) to the worker terminal 110.

  FIG. 14 shows a screen 1300 output by the operator terminal 110 in response to the depression of the search button 912 described above. The screen 1300 is configured of a search result display box 1301 and a selection button 1302.

  Among them, the search result display box 1301 is a display object for displaying the search result obtained by executing the plan support system 104 on the Internet 101 according to the content of the search input box 911. The selection button 1302 is a button for the planner to select the link destination displayed in the search result display box 1301. When the selection button 1302 is pressed, the temporary plan correction unit 111 of the plan support system 104 that receives this via the worker terminal 110 receives the external information link destination displayed in the corresponding search result display box 1301. Is displayed in the external information link destination input box 913 on the screen 900.

  Similarly, the above-mentioned planner operates the worker terminal 110 to input the link destination information of the internal information in the internal information link destination input box 915 of the screen 900, and presses the save button 916. The temporary plan correction unit 111 of the plan support system 104 that has received this pressing via the worker terminal 110 acquires link destination information of the internal information input in the internal information link destination input box 915.

  Thus, the temporary plan correction unit 111 of the plan support system 104 acquires information of each link destination of the external information and the internal information, associates these with the correction reason and the learning device, and registers them in the corresponding record of the correction reason DB 115 1410).

  Subsequently, the CPU 301 of the plan support system 104 inputs the order information, the correction reason, and the execution plan obtained in the steps up to here into the automatic planning system 109, and executes the learning process (1411). Details of this process in the automatic planning system 109 will be described later with reference to FIG.

Finally, the execution plan output unit 114 of the plan support system 104 outputs the execution plan determined in step 1410 to step 1411 to the management system or the like of the manufacturing line which is the internal system of the manufacturer (1412), and the flow is performed. finish.
--- Flow example 2--

  Next, the details of step 1402 in the above-described flow example 1 will be described. FIG. 15 is a diagram showing a flow example 2 of the plan support method in the present embodiment, and more specifically, is a diagram showing an operation flow of temporary plan creation in the automatic planning system 109.

  In this case, the CPU 301 of the planning support system 104 holds the external information and internal information read into the memory 302 in step 1401 described above and the correction reason DB 115 using the external / internal information processing unit 202 in the automatic planning system 109. The matching process with the existing planning reason is executed (1501).

  For example, the HTML source of weather information obtained from a predetermined Web site includes a character string "Typhoon comes in 2 days" by a search based on the external information link destination on the Internet 101 by the external information input unit 105 of the planning support system 104 It is assumed that In this case, the external / internal information processing unit 202 of the planning support system 104 specifies, among the respective planning reasons of the correction reason DB 115, the planning reason “Typhoon comes” including the keyword “Typhoon comes” by the matching process. Do.

  In addition, a natural language processing technology called morphological analysis is generally known. By using the present technology, the above character strings can be classified into parts of speech, and matching processing with each drafting reason of the correction reason DB 115 can be performed with respect to similar character strings. The external information and the reason for planning may be matched using such a method.

  In addition, regarding internal information, the external / internal information processing unit 202 sets, for example, the value of “sensor Y2” among the sensor data as internal information and the value of internal information to which the existing planning reason “sensor Y2 is abnormal” is attached. When it compares and it specifies that the value of the sensor data acquired now exceeds the applicable value of the existing planning reason, the said existing planning reason "sensor Y2 is abnormal" is specified as a result of a matching process.

  In addition, it is possible to use the discriminant analysis method to determine whether the value of “sensor Y2” is abnormal as described above, and in the case of abnormality, to match the existing planning reason “sensor Y2 is abnormal”. In this way, even if the value of “sensor Y2” falls below a certain value and becomes abnormal, it is possible to specify by matching the planning reason. Further, when perfect matching between the external information and the internal information and the existing planning reason can not be performed, the external / internal information processing unit 202 may cause the display of the terminal for worker 110 to display the fact that such matching can not be performed.

  Next, the CPU 301 of the planning support system 104 uses the learning device adopting unit 203 to specify, using the correction reason DB 115, a learning device corresponding to the planning reason specified in the matching of step 1501 described above (1502).

Subsequently, the CPU 301 of the plan support system 104 uses the temporary plan creation unit 204 to input order information to the learning device specified in step 1502 described above (1503). The CPU 301 of the plan support system 104 acquires a temporary plan output by the learning device to which the above order information is input, provides this to the process of step 1403 (1504), and ends the process.
--- Flow example 3--

  Next, the learning operation in the automatic planning system 109 will be described. FIG. 16 is a diagram showing a third flow example of the planning support method in the present embodiment, and more specifically, a learning flow of the automatic planning system 109. As shown in FIG.

In this case, the CPU 301 of the planning support system 104 can use the correction reason processing unit 206 of the automatic planning system 109 to read the above correction reason, collate the matching reason with the existing planning reason in the correction reason DB 115, and match them. In this case, a learning device linked to the existing planning reason is selected (1601). The matching method here is step 1501.
The same method as used in can be used.

  On the other hand, if the correction reason can not be matched with the existing planning reason, the CPU 301 of the planning support system 104 creates a new learning device by the learning unit 208 (1602). As already described above, the learning device is a planning algorithm configured including a learning model constructed using a machine learning technology such as a neural network. The learning algorithm, which is a planning algorithm, is a known (or similar) input to known (or similar) input data when known (or similar to) known input data is input. Or similar output as that of the known one, that is, one that outputs the planning result. Therefore, when creating a new learner, if a given learner is given the above-mentioned correction reason and order information, the parameter of the corresponding algorithm is output under a predetermined rule so that the corresponding execution plan is output. It is possible to assume a form of creating a new learning device by adjusting, etc.

  Using the execution plan processing unit 207, the CPU 301 of the plan support system 104 inputs the order information and the execution plan into the learning device selected in step 1601 or created in step 1602 (1603).

  The CPU 301 of the planning support system 104 uses the learning unit 208 to adjust the accuracy of the above-described learning device (1604).

  Here, the learning device and the accuracy adjustment of the learning device will be described. In this example, it is assumed that a neural network learning device and an error back propagation method are used. As the neural network learning device and the error back propagation method, existing techniques can be used.

  The inputs in this case are the order number, the delivery date, the quantity, and the type, and the outputs are the temporary manufacturing order and the machine number. Further, with regard to the temporary manufacturing order and the equipment number which are discrete values, the selected probability is calculated, and the order is determined in the descending order of the probability.

  The number of layers of the neural network learning device is three, the number of neurons in each layer is four in the first layer, ten in the second layer, and two in the third layer. In addition, activation functions between layers use sigmoid functions.

  There are commonly available machine learning software for these calculations, which can also be used. Although specific numerical values, item names and function names are listed as examples, other numerical values, item names, function names and the like may be handled. Also, in consideration of the production capacity of the production line, etc., one plan may be divided into production units which can be produced once. Further, for example, one order can be divided into a plurality of plans by dividing the order number 1 of FIG. 4 into a plurality of records “order number 1_1, order number 1_2, order number 1_3” in one quantity.

  In the above, the example applied to plan support was explained as an example of the present invention. The invention is also capable of other applications. For example, when there is a special situation such as power saving, the driver of the machine is made to record the reason why the work is different from usual and the link destination that can reflect the situation, and even if there is a special situation according to the present invention Automatic operation that can take account of Also, by displaying the probability selected for each planning reason on the screen 1200 of FIG. 13 showing the operation of the conventional correction reason selection button, it is possible to contribute to education for newcomers.

In addition, various modifications not limited to the above-described embodiment are also possible. For example, it is not limited to what comprises all the configurations described in the above-described embodiment. Also, part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of one embodiment. In addition, with respect to a part of the configuration of each embodiment, it is possible to add, delete, and replace other configurations.

  In addition, each configuration, function, processing unit, etc. described above may be realized by hardware, for example, by designing part or all of them with an integrated circuit. Further, each configuration, function, etc. described above may be realized by software by the processor interpreting and executing a program that realizes each function. Information such as a program, a table, and a file for realizing each function can be placed in a memory, a hard disk, a recording device such as an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD.

  Further, control lines and information lines indicate what is considered to be necessary for the description, and not all control lines and information lines in the product are necessarily shown. In practice, almost all configurations may be considered to be mutually connected.

  According to the present embodiment, it is possible to create a highly accurate and efficient plan based on the planning reason of the planner. In addition, it allows other planners to recognize the reason for planning, which is important information, and can be expected to be effective in preventing inappropriate planning revisions and planning errors.

  At least the following matters will be made clear by the description of the present specification. That is, in the planning support system of the present embodiment, the arithmetic device specifies the planning reason including the information matching the reference information by the collation, and at least one of the planning reason and the specified planning algorithm It is also possible to further execute a process of outputting the information of (1) to a predetermined device.

  According to this, it is possible to prevent the planner from also recognizing the reason for planning and making inappropriate plan correction and planning mistakes. As a result, accurate and efficient planning can be made based on the planning reason of the planner.

  In the plan support system according to the present embodiment, the storage device is information of a predetermined address indicating the acquisition destination of the reference information which is the basis of the planning reason for the planning reason and the planning algorithm corresponding to the planning reason. Are further stored in association with each other, and the arithmetic device acquires the latest reference information from the predetermined address which is the acquisition destination of the reference information at the time of the comparison, and the latest reference information and the respective plan The reason may be compared with the planning reason.

  According to this, external information viewed from the company to which the planner belongs, such as public information on the Internet (example: weather forecast etc.), and information managed by the business system etc. of the company to which the planner belongs (example: operating status of manufacturing equipment etc.) The reference information can be efficiently acquired based on the acquisition destination of the internal information such as information from the sensor observing the As a result, accurate and efficient planning can be made based on the planning reason of the planner.

  In the plan support system of the present embodiment, the computing device outputs the plan result acquired from the plan planning algorithm as a temporary plan to a predetermined device, and receives from the predetermined device the correction process by the next planner for the temporary plan. The temporary plan after the correction process may be output as a real plan to a predetermined device, and the real plan, the reason for the drafting, and the planning algorithm may be associated with each other and stored in the storage device. .

  According to this, it is possible for the planner to perform correction work on the provisional plan, and it is possible to compensate for the lack of planning accuracy in the planning algorithm. As a result, accurate and efficient planning can be made based on the planning reason of the planner.

  In the plan support system of the present embodiment, the computing device receives, from the predetermined device, a correction process of at least one of the planning reason and the predetermined address of the reference information by the next planner regarding the plan result. A process may be further performed in which at least one piece of information after the correction process and the actual plan are associated with each other and stored in a storage device.

  According to this, the planner can also correct the planning reason and the acquisition destination of the reference information, and consequently, it is possible to plan with high accuracy and efficiency based on the planning reason of the planner.

In the plan support system of the present embodiment, the arithmetic device is different in the reason for correction planning as the reason for planning corrected by the next planner and the existing planning reason specified with respect to the latest reference information obtained now In this case, the parameters of the specified planning algorithm are corrected so that the actual plan is output with the predetermined information on the next planned event and the correction planning reason as input, and the corrected planning algorithm is The process of storing in the storage device in association with the correction planning reason may be further executed.

  According to this, it is possible to improve the accuracy of the planning algorithm, and in turn, it is possible to plan with high accuracy and efficiency based on the reason for planning by the planner.

  In the plan support method of the present embodiment, the information processing system specifies the planning reason including the information matching the reference information by the collation, and at least one of the planning reason and the specified planning algorithm. A process of outputting information to a predetermined device may be further performed.

  In the plan support method according to the present embodiment, the information processing system in the storage device is an acquisition destination of the reference information that is the basis of the planning reason for the planning reason and the planning algorithm corresponding to the planning reason. The information of the predetermined address indicating the information is further stored in association, and the latest reference information is acquired from the predetermined address which is the acquisition destination of the reference information in the matching, and the latest reference information and the reason for planning each plan And may be compared.

  In the plan support method of the present embodiment, the information processing system outputs the plan result acquired from the plan planning algorithm as a provisional plan to a predetermined device, and the correction process by the next planner for the provisional plan is performed from the predetermined device The temporary plan after the correction process may be output as a real plan to a predetermined device, and the real plan may be further stored in the storage device in association with the planning reason and the planning algorithm.

  In the plan support method according to the present embodiment, the information processing system receives, from a predetermined apparatus, a correction process of at least one of the plan reason and the predetermined address of the reference information by the next planner regarding the plan result. A process of correlating at least one piece of information after the correction process with the actual plan and storing the same in the storage device may be further executed.

In the method for supporting planning according to the present embodiment, the information processing system includes a correction planning reason which is a planning reason corrected by the next planner and an existing planning reason which is specified with respect to the latest reference information obtained now. If different, the parameters of the specified planning algorithm are corrected so that the actual plan is output based on the predetermined information on the next planned event and the correction planning reason, and the corrected planning algorithm is A process of storing in the storage device in association with the correction planning reason may be further executed.

101 Internet 102 Order Information DB
103 Internal Information DB
104 plan support system 105 external information input unit 106 order information input unit 107 internal information input unit 108 learner selection unit 109 automatic planning system 110 terminal for worker 111 temporary plan correction unit 112 temporary plan DB
113 Execution plan DB
114 Execution plan output unit 115 Correction reason DB
201 temporary plan creation device 202 external / internal information processing unit 203 learning device adoption unit 204 temporary plan creation unit 205 learning device 206 correction reason processing unit 207 execution plan processing unit 208 learning unit 301 CPU (computing device)
302 Memory 303 Interface 304 Network Interface 305 Keyboard 306 Display 307 Mouse 308 Storage Device 310 Program 311 OS
900 input screen 901 planner number input box 902 temporary planning display box 903 temporary planning correction box 906 temporary planning reason display box 907 learner selection button 908 execution plan display box 910 correction reason input box 913 external information link destination input box 915 internal Information link destination input box 1000 screen 1100 screen 1200 screen 1300 screen

Claims (12)

And drafting reasons planning already planned, a storage device for storing in association with planning algorithm is configured to include a constructed learning model using a machine learning technique used in the program for,
The reference information acquired from the predetermined address on the predetermined network is compared with the planning reason of each plan to perform matching processing, and the planning reason including the information matching the reference information specified in the matching processing identify the planning algorithm that is associated with planning reasons, with respect to the planning algorithm, and obtains the planning results by entering predetermined information on Imatsugi plan events, and outputs the planning results to a predetermined device A computing device,
Planning support system characterized by including. The arithmetic device is
The matching is performed to specify a planning reason including information matching the reference information, and to further execute processing of outputting at least one of the planning reason and the specified planning algorithm to a predetermined device. ,
The plan support system according to claim 1, wherein the plan support system comprises: The storage device is
Information on a predetermined address indicating the acquisition destination of the reference information that is the basis of the planning reason is further stored in association with the planning reason and the planning algorithm corresponding to the planning reason,
The arithmetic device is
At the time of the matching, the latest reference information is acquired from a predetermined address from which the reference information is acquired, and the latest reference information is collated with the planning reason of each plan.
The plan support system according to claim 1, wherein the plan support system comprises: The arithmetic device is
The plan result acquired from the planning algorithm is output as a provisional plan to a predetermined device, the correction process by the next planner for the provisional plan is received from the predetermined device, and the provisional plan after the correction process is regarded as an actual plan. , And the processing for correlating the actual planning with the planning reason and the planning algorithm and storing them in the storage device.
The planning support system according to claim 3, characterized in that: The arithmetic device is
With regard to the result of the plan, the process of correcting the information on at least one of the reason for planning and the predetermined address of the reference information by the next planner is received from the predetermined device, and at least any information after the correction process and the actual It further executes processing of associating the plan and storing it in the storage device.
The planning support system according to claim 4, characterized in that: The arithmetic device is
If there is a difference between the reason for correction that is the reason for the correction made by the next planner and the existing reason for planning that is identified with respect to the latest reference information that is obtained now, Modifying the parameters of the specified planning algorithm so as to output the actual plan with the reason as an input, and storing the corrected planning algorithm in the storage device in association with the correction planning reason Further to be executed,
The planning support system according to claim 5, characterized in that An information processing system including a storage device that stores a planning reason of a planned plan and a planning algorithm configured including a learning model constructed using a machine learning technique used for the plan in association with each other.
The reference information acquired from the predetermined address on the predetermined network is compared with the planning reason of each plan to perform matching processing, and the planning reason including the information matching the reference information specified in the matching processing identify the planning algorithm that is associated with planning reasons, with respect to the planning algorithm, and obtains the planning results by entering predetermined information on Imatsugi plan events, and outputs the planning results to a predetermined device ,
Planning support method characterized by The information processing system
A process of specifying a planning reason including information matching the reference information by the collation, and outputting at least one of information of the planning reason and the specified planning algorithm to a predetermined device is further executed.
The plan support method according to claim 7, characterized in that: The information processing system
The storage device further stores information of a predetermined address indicating the acquisition destination of the reference information that is the basis of the planning reason in association with the planning reason and the planning algorithm associated with the planning reason,
At the time of the collation, the latest reference information is acquired from a predetermined address from which the reference information is acquired, and the latest reference information is collated with the planning reason of each plan.
The plan support method according to claim 7, characterized in that: The information processing system
The plan result acquired from the planning algorithm is output as a provisional plan to a predetermined device, the correction process by the next planner for the provisional plan is received from the predetermined device, and the provisional plan after the correction process is regarded as an actual plan. , And the processing for correlating the actual planning with the planning reason and the planning algorithm and storing them in the storage device.
The plan support method according to claim 9, characterized in that: The information processing system
With regard to the result of the plan, the process of correcting the information on at least one of the reason for planning and the predetermined address of the reference information by the next planner is received from the predetermined device, and at least any information after the correction process and the actual Further execute processing of associating the plan and storing it in the storage device;
The plan support method according to claim 10, characterized in that: The information processing system
If there is a difference between the reason for correction that is the reason for the correction made by the next planner and the existing reason for planning that is identified with respect to the latest reference information that is obtained now, Modifying the parameters of the specified planning algorithm so as to output the actual plan with the reason as an input, and storing the corrected planning algorithm in the storage device in association with the correction planning reason Do more
The plan support method according to claim 11, characterized in that:

Images