JP6886101B2 - Information processing equipment, information processing methods, programs - Google Patents

Description

The present invention relates to an information processing device, an information processing method, and a program.

In allocation plans such as assortment plans, production plans, and delivery plans, allocations when allocating objects (orders, tasks, delivery orders, etc.) to allocation destinations (raw fabrics, machines, vehicles, etc.) Decide which one.

The allocation method is determined so that the combination of the allocation target and the allocation destination, the arrangement, the delivery date, and other constraints are satisfied, and the evaluation of the yield rate, operating rate, loading rate, etc. is improved.

When the allocation plan is done manually, the person in charge of planning considers the constraints and evaluation. Some of these constraints and evaluations are consciously considered by the planner, while others are unconsciously considered.

On the other hand, when the allocation plan is automatically performed, it is necessary to formulate the constraint conditions and evaluation knowledge possessed by the planner.

Of these, the knowledge that the person in charge of planning is aware of can be verbalized by hearing from the person in charge. However, there is knowledge that is difficult to formulate even if it can be verbalized, such as "planning with a" good "balance between delivery time and efficiency."

Furthermore, knowledge that the planner is not aware of is verbalized and formulated by trial and error, such as prototyping an algorithm for automatic planning, presenting the allocation plan results, and adjusting the algorithm from the evaluation comments of the planner. Since it is necessary to go, it is difficult and time-consuming to implement.

Therefore, Patent Document 1 describes a method of finding an optimum production rule (dispatching rule) by using machine learning for a production plan, which is one of the allocation plans, and automatically creating a production plan based on the production rule. ing.

Japanese Unexamined Patent Publication No. 2004-94900

The above-mentioned Patent Document 1 discloses the following method.

First, a planning simulator that performs simulation based on predetermined production rules (dispatching rules) is prepared. Next, simulations are obtained for various production conditions (production process / order, etc.) based on a plurality of production rules. The simulation results are evaluated based on given evaluation indexes such as inventory amount and machine operating rate. The combination of each production status and the optimum (maximum evaluation value) production rule is accumulated as teacher data. By learning with a neural network (NN) using the accumulated teacher data, a network with the production status as Input and the optimum production rule as Output is completed.

However, the method of Patent Document 1 cannot be formulated like "optimal" based on human senses because it can be used only when the evaluation index is given, that is, the criterion of "optimal" is clear. Not available if.

Furthermore, since the simulation is performed based on the determined production rules, the production rules are not dynamically adjusted depending on the state during the simulation, so the production rules should be dynamically adjusted. There is concern that it may have overlooked the possibility that higher-rated results were obtained.

Therefore, an object of the present invention is to provide a mechanism capable of formulating an allocation plan according to the planning knowledge even when the planning knowledge possessed by the person in charge of planning is difficult to formulate.

The information processing apparatus of the present invention has an allocation plan creating means for creating an allocation plan for allocating an allocation target to an allocation destination according to a constraint condition, and a change instruction or a new allocation target for the allocation plan created by the allocation plan creating means. An instruction receiving means for receiving an object allocation instruction, an instruction content specifying means for specifying which of the predetermined candidates corresponds to a change instruction or a new allocation instruction received by the instruction receiving means, and the allocation plan. A teacher data extraction means that extracts a set of an allocation plan created by the creation means and an instruction content specified by the instruction content specifying means as teacher data, and a teacher data extraction means.
It is characterized by having.

Further, in the information processing method of the present invention, the allocation plan creation means of the information processing apparatus creates an allocation plan for allocating the allocation target to the allocation destination according to the constraint conditions, and an instruction reception of the information processing apparatus. The instruction receiving process in which the means receives a change instruction for the allocation plan created by the allocation plan creating process or an allocation instruction for a new allocation target object, and the instruction content specifying means of the information processing apparatus accepts in the instruction receiving process. The instruction content specifying process for specifying which of the predetermined candidates corresponds to the change instruction or the new allocation instruction, and the teacher data extraction means of the information processing apparatus are created by the allocation plan creation process. It is characterized by including a teacher data extraction step of extracting a set of an allocation plan and an instruction content specified by the instruction content specifying step as teacher data.

Further, the program of the present invention is a program that can be executed in the information processing apparatus, and the allocation plan creating means for creating an allocation plan for allocating the allocation target to the allocation destination in accordance with the constraint conditions, and the above-mentioned. Predetermined candidates for the instruction receiving means for receiving the change instruction for the allocation plan created by the allocation plan creating means or the allocation instruction for the new allocation target, and the change instruction or the new allocation instruction received by the instruction receiving means. Teacher data extraction that extracts a set of the instruction content specifying means for specifying which of the above is applicable, the allocation plan created by the allocation plan creating means, and the instruction content specified by the instruction content specifying means as teacher data. A program to function as a means.

According to the present invention, even when it is difficult to formulate the planning knowledge possessed by the person in charge of planning, it is possible to formulate an allocation plan according to the planning knowledge.

It is a drawing which shows the outline of the assortment planning problem which is 1 Example of this invention. It is a system block diagram which shows an example of the structure of the allocation planning system of this invention. It is a block diagram which shows the hardware configuration of the information processing apparatus applicable to the management server 103 or the client terminal 101 shown in FIG. It is a figure which shows an example of the functional structure of the management server 103 and the client 101 shown in FIG. It is a flowchart which shows the outline of the teacher data extraction processing in the assortment planning system in this invention. It is a figure which shows an example of the screen of the intention input in the input / display control unit. It is a figure which listed the partial allocation candidate with respect to the assortment result data. It is a figure which shows an example of the evaluation of the partial allocation candidate with respect to the intention result data. It is a drawing which shows the neural network used for the machine learning processing in the assortment planning system in this invention. It is a flowchart which shows the outline of the arrangement plan making process in the arrangement planning system in this invention. It is a flowchart which shows the detail of the step S104 partial allocation evaluation process of FIG. It is a figure which shows the state which initialized the evaluation value of a partial allocation candidate. It is a figure which shows the state which tentatively allocated based on the partial allocation candidate of the first rank with respect to the assortment result. It is a figure which shows the result of the temporary allocation according to the partial allocation candidate with the highest evaluation value, and the allocation simulation after that. It is a figure which shows the evaluation value update of a partial allocation candidate. It is a figure which shows an example of the data structure of an order table. It is a figure which shows an example of the data structure of the original fabric table. It is a figure which shows an example of the data structure of the assortment result image and the assortment result table. It is a figure which shows an example of the data of the intention result table. It is a figure which shows an example of the data structure of a teacher data table.

Hereinafter, an example of the embodiment of the present invention will be described with reference to the drawings.

In the explanation, the assortment planning problem, which is one of the allocation planning problems, is the subject. FIG. 1 is a diagram showing an outline of an assortment planning problem, which is one of the allocation planning problems.

The assortment planning problem is the problem of arranging the order to be allocated to the original fabric to which it is allocated. Arrangement is performed so that the constraints are satisfied and the evaluation is good. Constraints include restrictions on the combination of order and raw fabric (matching the raw fabric of the material specified in the order, etc.), order constraints (matching all orders, etc.), and placement constraints (in the short side direction). There are various restrictions such as not being able to place more than 4 orders). In addition, various evaluation axes can be considered for the evaluation, such as the yield rate (the ratio of the total area of the combined orders to the original fabric area) and the number of setup changes. It should be noted that there may be some constraints and evaluations that cannot be verbalized or formulated.

Although the subject of this embodiment is the assortment planning problem, it can also be applied to other allocation planning problems.

FIG. 2 is a diagram showing an example of the system configuration of the allocation planning system according to the present invention. The management server 103 in the figure is a device that functions as an information processing device of the present invention, and when it receives an allocation plan creation instruction from client devices 101 and 102, it performs an allocation plan creation process described later. Reference numerals 101 and 102 are client devices, which are terminal devices used by the user to give an allocation plan creation instruction to the management server 103. The 103 is a network such as a LAN (Local Area Network) that connects the management server 103 and the client devices 101 and 102 so as to be able to communicate with each other.

Next, an example of the hardware configuration of the information processing device applicable to the management server 103 of FIG. 1 will be described with reference to FIG.

The CPU 201 comprehensively controls each device and controller described later connected to the system bus 204. Further, the ROM 202 or the external memory 211 is necessary to realize the functions executed by the BIOS (Basic Input / Output System) and the operating system program (hereinafter, OS), which are the control programs of the CPU 201, and each server or each PC. Various programs described later are stored. The RAM 203 functions as a main memory, a work area, and the like of the CPU 201.

The CPU 201 realizes various operations by loading a program or the like necessary for executing a process into the RAM 203 and executing the program. Further, the input controller (input C) 205 controls the input from a pointing device such as a keyboard 209 or a mouse (not shown). The video controller (VC) 206 controls the display on a display device such as the display device 210. The display device is, for example, a CRT display, a liquid crystal display, or the like. These are used by the administrator as needed. It is not directly related to the present invention.

The memory controller (MC) 207 can be used in a hard disk (HD), floppy (registered trademark) disk, or PCMCIA card slot for storing boot programs, browser software, various applications, font data, user files, editing files, various data, and the like. Controls access to external memory 211 such as CompactFlash® memory connected via an adapter.

The communication I / F controller (communication I / FC) 208 connects and communicates with an external device via a network, and executes communication control processing on the network. For example, Internet communication using TCP / IP is possible.

The CPU 201 enables display on the display device 210 by, for example, executing an outline font expansion (rasterization) process in the display information area in the RAM 203. Further, the CPU 201 enables a user instruction with a mouse cursor or the like (not shown) on the display device 210.

The program used to execute various processes of the management server of the present invention is recorded in the external memory 211, and is executed by the CPU 201 by being loaded into the RAM 203 as needed. Further, the definition file and various information tables used by the program according to the present invention are stored in the external memory 211. The above is an explanation of an example of the hardware configuration of the information processing device applicable to the management server 103.

FIG. 4 is a schematic diagram showing an example of the functional configurations of the client 101 and the management server 103. As shown in FIG. 4, the management server 103 includes a data management unit 301, a transmission control unit 302, a reception control unit 303, an assortment plan creation unit 304, a teacher data extraction unit 305, and a machine learning unit 306. Further, the client 101 includes a transmission control unit 311, a reception control unit 312, and an input / display control unit 313.

The data management unit 301 manages order data, original fabric data, assortment result data, intention result data, teacher data, and assortment rule data, which are data required for assortment planning, teacher data extraction, and machine learning. The order table 301-1, the original fabric table 301-2, the assortment result table 301-3, the intention result table 301-4, the teacher data table 301-5, and the assortment rule table 301-6 are managed. It is a functional part. The structure of these data tables will be described later.

The transmission control unit 302 is a functional unit that transmits necessary data from the data managed by the data management unit 301 to the client 101. The reception control unit 303 is a functional unit that stores the data received from the client 101 in the data management unit 301.

The assortment planning unit 304 arranges the order (allocation target) of the order table 301-1 with the original fabric (allocation destination) of the original fabric table 301-2 according to the assortment rule (constraint condition) of the assortment rule table 301-6. It is a functional part that formulates an assortment plan (allocation plan).

The teacher data extraction unit 305 is a functional unit that extracts as teacher data the allocation elements to be ruled as an assortment rule from the intention result of the intention result table 301-4. The machine learning unit 306 is a functional unit that performs machine learning according to the teacher data in the teacher data table 301-5 and obtains an assortment rule.

The transmission control unit 311 is a functional unit that transmits necessary data from the data input by the input / display control unit 313 to the server 103. The reception control unit 312 is a functional unit that transfers the data received from the server 103 to the input / display control unit 313. The input / display control unit 313 is a function unit that displays the assortment result on the screen, which is delivered from the reception control unit 312, receives an intention input such as order allocation and change, and delivers the result to the transmission control unit 311. ..

Next, the data structure of the order table 301-1 of FIG. 4 will be described with reference to FIG. As shown in FIG. 16, the order table 301-1 is configured to have data items such as an order ID, a long side, a short side, an order quantity, and a cutting delivery date.

Next, the data structure of the original fabric table 301-2 of FIG. 4 will be described with reference to FIG. As shown in FIG. 17, the original fabric table 301-2 is configured to have data items such as an original fabric ID, a long side, and a short side.

Next, the data structure of the assortment result table 301-3 of FIG. 4 will be described with reference to FIG. As shown in FIG. 18, the assortment result table 301-3 shows the assortment ID, the original fabric ID, and the partial allocation No. , Order ID, long side position, short side position, long side coordinates, short side coordinates.

Partial allocation No. Is a serial number for identifying the order placed on the original fabric. The long-side position and the short-side position indicate the arrangement according to the partial allocation candidate arrangement condition, and indicate how they are in contact with the already arranged partial allocation or the original fabric end. The partial allocation candidate placement conditions will be described later. The long side coordinates and the short side coordinates indicate the coordinates of the upper left corner of the order when the upper left corner of the original fabric is set to 0.

Next, the data structure of the intention result table 301-4 of FIG. 4 will be described with reference to FIG. As shown in FIG. 19, the intention result table 301-4 is configured to have some items (order ID, long side coordinates, short side coordinates) for one partial allocation of the assortment result. There is.
Hereinafter, the details of each function will be described along with the processing flow.

The outline of the intention input will be described with reference to FIG. The intention input means that a person makes an additional placement of an order (assignment instruction for a new allocation target) or a placement change (change instruction for an allocation plan) with respect to the assortment result data. FIG. 6 is a diagram showing an example of an intention input screen in the input / display control unit 313. On the intention input screen, a list of original fabrics, a list of orders, a list of details of the assortment result, and a plan layout of the assortment result data are displayed.

When making an intention, in the case of additionally arranging a new order, a desired order is selected from the order list and arranged at a desired position in the plan layout. In the case of rearranging the placed orders, the placement position of the partial allocation on the plan layout is changed.

Here, it should be noted that since the intention is made manually, the position does not necessarily match the partial allocation candidate placement condition (details will be described later).

The outline of the teacher data extraction process performed by the CPU 201 of the management server 103 will be described with reference to FIG. The computer program for causing the CPU 201 to perform this process is stored in the external memory 211. If necessary, the CPU 201 loads the computer program into the RAM 203 and performs this process according to the control of the loaded computer program. become.

First, the CPU 201 uses various data used for the teacher data managed by the data management unit 301.

(Step S001). The data acquired here is the intention result data stored in the intention result table 301-4.

Subsequent processing (steps S003 to S007) is repeated for each intention result data acquired in step S001 (step S002).

In the iterative process, the CPU 201 first lists the partial allocations that match the partial allocation candidate placement conditions (step S003).

Here, the partial allocation candidate placement conditions will be described. The partial allocation candidate placement condition is a condition for limiting the position where the partial allocation is placed with respect to the assortment result, and is a concept required when constructing the assortment result by combining the partial allocation in the assortment planning. is there. The reason why it is necessary will be described later.

In this embodiment, the partial allocation candidate placement condition is predetermined as "both the long side coordinates and the short side coordinates are positions that are in contact with the already placed partial allocation or the original fabric end". It is important to design the partial allocation candidate placement condition so as to be consistent with the assortment rule to be learned and the assortment plan planning algorithm, and this embodiment is also designed accordingly.

FIG. 7 shows an image in which partial allocation candidates that match the partial allocation candidate placement conditions are listed with respect to the assortment result data.

After the partial allocation candidate enumeration in step S003 is completed, the difference from the intention result data is evaluated for each partial allocation candidate (steps S004 to S006).

As an example of the difference evaluation method, FIG. 8 shows an example in which the distance between the partial allocation candidate and the placement position of the intention result data is evaluated as an evaluation value.

Next, the partial allocation candidate with the highest evaluation is used as teacher data (step S007). That is, the difference between the content of the change instruction or the allocation instruction and the predetermined candidate is evaluated for each candidate, and the candidate having the highest evaluation value is specified as the instruction content and used as teacher data. In the example of FIG. 8, when the long side position is arranged so as to be in contact with the right side of the left end of the original fabric and the short side position is arranged so as to be in contact with the lower side of the partial allocation 2, the distance is the minimum (71.2). Extract as data.

As shown in FIG. 20, the data structure of the teacher data table 301-5 is such that the teacher data table 301-5 is composed of a combination of the allocation state and the partial allocation, and the allocation state is the data including the information of the assortment result data. The partial allocation has the same data items as the partial allocation candidates.

After the iterative execution is completed for all the intention result data (step S008), the teacher data is output (step S009). In this step, the created teacher data is output (transmitted) to the server 103. Then, the server 103 that has received the teacher data stores the teacher data in the teacher data table 301-5. The above is the description of the teacher data extraction process of FIG.

Next, the outline of the machine learning process will be described with reference to FIG.

In the machine learning process in this embodiment, assortment rules are extracted from the teacher data. Here, the teacher data is a "assignment state and a set of partial allocations selected by a person for the allocation state", and the assortment rule obtained as a result of learning is the likelihood of the partial allocation candidate with respect to the allocation state ( Selection probability). By setting the rule target of the assortment rule to one partial allocation instead of the entire assortment result, it is possible to improve the learning efficiency by reducing the variables (degrees of freedom) of the learning target. Further, in the teacher data extraction process of FIG. 5, the teacher data is limited to those that match the partial allocation candidate placement condition, which also contributes to reducing the variables (degrees of freedom) to be learned. The means for formulating an assortment plan using this assortment rule will be described later.

In general, machine learning refers to iterative learning from data and finding patterns hidden in it, and there are various techniques such as genetic algorithm (GA), support vector machine (SVM), and neural network (NN). To do. FIG. 9 shows an example of a model when NN is adopted as a technique.

Next, with reference to FIG. 10, an outline of the arrangement planning process performed by the CPU 201 of the management server 103 will be described. The computer program for causing the CPU 201 to perform this process is stored in the external memory 211. If necessary, the CPU 201 loads the computer program into the RAM 203 and performs this process according to the control of the loaded computer program. become.

In the assortment planning process, the CPU 201 first acquires various data used for the teacher data managed by the data management unit 301 (step S101). The data acquired here is stored in the order table 301-1. Order data, raw fabric data stored in raw fabric table 301-2, assortment result data (progress) stored in assortment result table 301-3, assortment rule stored in assortment rule table 301-6 Data.

For the data acquired in step S101, the subsequent processes (steps S103 to S106) are repeated until the assortment is completed (step S102).

Completion of the assortment is determined by conditions such as all orders have been allocated, the remaining space of the original fabric is exhausted, and the calculation time reaches a given calculation time upper limit.

In the iteration, the CPU 201 performs a partial allocation evaluation for all the partial allocation candidates that match the partial allocation candidate placement conditions (steps S103 to S105). Details of the partial allocation evaluation will be described later with reference to FIG.

Next, allocation (arrangement) is performed according to the partial allocation candidate with the highest evaluation (step S106). Since the partial allocation candidate has information on how to contact the already arranged partial allocation or the original fabric end, it is possible to specify the position (long side coordinate, short side coordinate) to be arranged.

After the assortment is completed (step S107), the assortment result data is output (step S108). In this step, the created assortment result data is output (transmitted) to the server 103. Then, the server 103 that has received the assortment result data stores the assortment result data in the assortment result table 301-3. The above is the description of the assortment planning process of FIG.

Next, with reference to FIG. 11, the details of the partial allocation evaluation process in step S104 of FIG. 10 will be described. The computer program for causing the CPU 201 to perform this process is stored in the external memory 211. If necessary, the CPU 201 loads the computer program into the RAM 203 and performs this process according to the control of the loaded computer program. become.

The partial allocation evaluation process calculates an evaluation value for obtaining the optimum partial allocation for the intermediate state of the assortment result. In calculating the evaluation value, the assortment rule data in the assortment rule table 301-6 is used. According to the assortment rule, the likelihood of partial allocation can be calculated for the allocation status (including the intermediate state of the assortment plan). However, since it is a local evaluation for the partial allocation to the last, even if the allocation is repeated according to the partial allocation with the highest likelihood, the final assortment result is not always suitable. Therefore, in the partial allocation evaluation process, a method is adopted in which the final assortment result is made suitable by pre-reading and evaluating the allocation while using the assortment rule data.

In the partial allocation evaluation process, the CPU 201 first initializes the evaluation values of the partial allocation candidates by evaluation according to the assortment rule (step S201).

FIG. 12 shows an example of a state in which the evaluation values of the partial allocation candidates are initialized by the evaluation based on the likelihood according to the assortment rule. The assortment result details are in the middle of the assortment plan, and two orders A01 are assigned to the original fabric. The partial allocation candidate list is a list of partial allocation candidates in the relevant state, and lists two orders and the long side position and short side position adjacent to the original fabric end. For the evaluation value of the partial allocation candidate list, the value obtained by evaluating the partial allocation candidate according to the assortment rule is set as the initial value. Also, the number of temporary allocation simulations is initialized to 0. The method of using the number of temporary allocation simulations will be described later.

Next, the subsequent processing (steps S203 to S205) is repeated until the given calculation time is exceeded (step S202).

In the iterative process, the CPU 201 first performs provisional allocation according to the partial allocation candidate having the highest evaluation (step S203). In the partial allocation candidate list of FIG. 12, the partial allocation candidate of rank 1 has the highest evaluation value, and provisional allocation is performed based on the highest evaluation value. FIG. 13 shows an intermediate state of the assortment result after the provisional allocation.

Since the number of temporary allocation simulations is all 0 at the first iteration, the partial allocation candidate with the highest evaluation was selected, but after the first time, the number of temporary allocation simulations is incorporated into the selection criteria of the partial allocation candidates as a bias term, and temporary allocation is performed. Priority should be given to partial allocation candidates with a small number of times.

UCB1 (Upper Confidence Bound) and the like are known as a method of incorporating the bias term.

Next, in step S204, a temporary allocation simulation is performed for the state of the assortment result in which the provisional allocation is performed until the assortment is completed. Since this temporary allocation simulation is performed to measure the effectiveness of the temporary allocation performed in step S203, it is required to be performed at high speed. As an example of a specific method, partial allocation is selected based on the selection probability obtained by the assortment rule, provisional allocation is performed, and this is repeated to create an assortment result. FIG. 14 shows an example of the assortment result obtained by the provisional allocation simulation.

The assortment result created by the provisional allocation simulation is evaluated, and the evaluation value of the partial allocation candidate selected in step S203 is updated based on the evaluation (step S205).

The evaluation of the provisional allocation simulation result may use a formalizable evaluation axis such as the yield rate, or may be evaluated based on the result of machine learning by accumulating the evaluation by the person in charge of planning. FIG. 15 shows a state in which the evaluation value of the partial allocation candidate is updated. The above is the description of the partial allocation evaluation process of FIG.

Although the embodiments have been described above, the present invention can be implemented as, for example, a system, an apparatus, a method, a program, a recording medium, or the like. Specifically, it may be applied to a system composed of a plurality of devices, or may be applied to a device composed of one device.

Further, the program in the present invention is a program in which a computer can execute the processing methods of the flowcharts shown in FIGS. 5, 10 and 11, and the storage medium of the present invention uses the processing methods of FIGS. 5, 10 and 11. A program that can be executed by a computer is stored. The program in the present invention may be a program for each processing method of each device of FIGS. 5, 10 and 11.

As described above, a recording medium on which a program that realizes the functions of the above-described embodiment is recorded is supplied to the system or device, and the computer (or CPU or MPU) of the system or device stores the program in the recording medium. Needless to say, the object of the present invention can be achieved by reading and executing.

In this case, the program itself read from the recording medium realizes the novel function of the present invention, and the recording medium on which the program is recorded constitutes the present invention.

Recording media for supplying programs include, for example, flexible disks, hard disks, optical disks, magneto-optical disks, CD-ROMs, CD-Rs, DVD-ROMs, magnetic tapes, non-volatile memory cards, ROMs, EEPROMs, and silicon. A disk or the like can be used.

Further, by executing the program read by the computer, not only the function of the above-described embodiment is realized, but also the OS (operating system) or the like running on the computer is actually operated based on the instruction of the program. Needless to say, there are cases where a part or all of the processing is performed and the processing realizes the functions of the above-described embodiment.

Further, after the program read from the recording medium is written in the memory provided in the function expansion board inserted in the computer or the function expansion unit connected to the computer, the function expansion board is based on the instruction of the program code. It goes without saying that there is a case where the CPU or the like provided in the function expansion unit performs a part or all of the actual processing, and the function of the above-described embodiment is realized by the processing.

Further, the present invention may be applied to a system composed of a plurality of devices or a device composed of one device. It goes without saying that the present invention can also be applied when it is achieved by supplying a program to a system or an apparatus. In this case, by reading the recording medium in which the program for achieving the present invention is stored into the system or the device, the system or the device can enjoy the effect of the present invention.

Further, by downloading and reading a program for achieving the present invention from a server, database, or the like on the network by a communication program, the system or device can enjoy the effect of the present invention. It should be noted that all the configurations in which each of the above-described embodiments and modifications thereof are combined are also included in the present invention.

101 Management server 102, 103 Client terminal

Claims (6)

An instruction receiving means for receiving an instruction to change the allocation plan or an instruction to allocate a new allocation object to the allocation destination for the allocation plan for allocating the allocation target to the allocation destination.
With respect to the change instruction or the new allocation instruction received by the instruction receiving means, the instruction content specifying means for specifying which of the predetermined candidates corresponds to the instruction.
A teacher data generating means for generating teacher data in which the allocation plan and the instruction content specified by the instruction content specifying means are combined, and
An information processing device characterized by being equipped with. An evaluation means for evaluating the difference between the content of the change instruction or the new allocation instruction and the predetermined candidate is further provided for each candidate.
The information processing apparatus according to claim 1 , wherein the instruction content specifying means identifies a candidate having the highest evaluation value by the evaluation means as the instruction content. Further provided with a display means for displaying the allocation plan,
The instruction receiving means is characterized in that it receives a change instruction or a new allocation instruction on the screen on which the allocation plan is displayed.
The information processing apparatus according to claim 1 or 2 , wherein the display means further adds information indicating the predetermined candidate to the allocation plan and displays the information. Further provided with an allocation rule generation means for generating an allocation rule using the result of learning the teacher data generated by the teacher data generation means.
The information processing apparatus according to any one of claims 1 to 3 , wherein the allocation plan is an allocation plan created based on the allocation rule generated by the allocation rule generating means. An instruction receiving process in which the instruction receiving means of the information processing device receives an instruction to change the allocation plan or an instruction to allocate a new allocation target to the allocation destination with respect to the allocation plan for allocating the allocation target to the allocation destination.
An instruction content specifying step for specifying which of the predetermined candidates corresponds to the change instruction or the new allocation instruction received by the instruction receiving process by the instruction content specifying means of the information processing device.
Tutor data generating means of the information processing apparatus, and the teacher data generating step of generating teacher data to set the instruction content specified by the allocation plan and the instruction content specifying means,
An information processing method characterized by being provided with. Computer,
An instruction receiving means for receiving an instruction to change the allocation plan or an instruction to allocate a new allocation object to the allocation destination for the allocation plan for allocating the allocation target to the allocation destination.
With respect to the change instruction or the new allocation instruction received by the instruction receiving means, the instruction content specifying means for specifying which of the predetermined candidates corresponds to the instruction.
A program for functioning as a teacher data generation means for generating teacher data in which the allocation plan and the instruction content specified by the instruction content specifying means are combined.

Images