JP2024010849A - Plan creation device and plan creation method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a plan creation device and a plan creation method capable of realizing continuously stable production by performing automatic update at appropriate timing without updating devices haphazardly.

SOLUTION: A plan creation device with a manual correction plan obtained by correction of an automatically generated plan by a plan creator as a final plan includes: a storage device for storing the automatically generated plan created by the automatic plan design processing portion and a manual correction plan obtained by correction of the automatically generated plan by the plan creator; a central processing unit having an evaluation processing portion for evaluating multiple evaluation indexes that are predetermined for the automatically generated plan and the manual correction plan, respectively, a detection processing portion for obtaining a difference between an evaluation result of the automatically generated plan and an evaluation result of the manual correction plan, and a proposal processing portion for recommending update timing of adjustment of weight of an evaluation index for minimizing the difference when the difference is larger than a predetermined threshold; and an outputting device for presenting adjustment of weight of an evaluation index recommended by the proposal processing portion to the plan creator.

SELECTED DRAWING: Figure 1

COPYRIGHT: (C)2024,JPO&INPIT

Description

The present invention relates to a plan creation device and a plan creation method that employs, as a final plan, a manually revised plan obtained by modifying an automatically generated plan by a planner.

There are many situations where advance planning is important, such as manufacturing products and operating and managing large-scale systems. When creating such a plan, the objective functions of the plan, such as maximizing production volume, maximizing equipment utilization rate, and minimizing the number of workers, must be determined while observing constraints such as time, space, equipment, human resources, etc. It is necessary to create a plan that satisfies the following. Creating such a plan manually would take too much time, so computers are often used.

On the other hand, in real environments, multiple constraints and multiple objective functions are often complex, so there are cases where it is difficult to accurately define all constraints and objective functions and input them into a computer. In such cases, the accuracy of planning can be improved by defining constraints and objective functions to the extent possible and combining them with logic based on empirical rules. However, it is extremely difficult to define constraints and objective functions that perfectly simulate the real environment on a computer, and logic based on empirical rules lacks versatility and scalability, so planning can be done using the plans output by the computer as they are. It is difficult to achieve something that satisfies the creator.

For this reason, there is a method that measures the ambiguous priorities of multiple constraints and objective functions based on the plan output by the computer and revised by the planner, and then reflects those priorities in the creation of the next plan. Proposed.

For example, according to Patent Document 1, "the plan creation unit 112 , using the plan condition data 121 and modified know-how data 126 to 129 read from the storage device 120, a new plan is created so that the new plan satisfies the constraint conditions and the objective function and conforms to one of the groups included in the modified know-how data. Determine decision variables for the plan.Here, the modification know-how data includes an anti-pattern 127 that indicates the tendency of work patterns of multiple plans before modification, and multiple modifications made by the plan creator to the multiple plans before modification. It includes a plurality of groups in which the reference pattern 128 indicating the trend of the work pattern of the revised plan, and the modification probabilities 126 which are the statistics of the plan-related information and decision variables of the plurality of unrevised plans are determined. I am proposing that.

JP2021-071771A

According to Patent Document 1, it is possible to create a plan with higher accuracy by reflecting the modification know-how extracted from the plan before modification and the plan after modification by the planner.

However, people's corrections are not always correct, and there may be other reasons, for example. Further, in Patent Document 1, correction know-how is extracted from the difference between the plan before correction and the plan after correction, but it is difficult to extract policy changes such as changes in the priority of each index from the difference.

Based on the above, the present invention provides a planning device and a planning method that make it possible to achieve sustainable and stable production by automatically updating the planning device at an appropriate time without unnecessarily updating the planning device. The purpose is to provide

According to the present invention, there is provided a plan creation device that uses, as a final plan, a manually revised plan obtained by modifying an automatically generated plan by a planner, wherein an automatically generated plan created by an automatic planning processing unit and an automatically generated a storage device that stores manually revised plans modified by the plan creator; an evaluation processing unit that evaluates a plurality of predetermined evaluation indicators for the automatically generated plans and manually revised plans; and an evaluation processing unit that stores the evaluation results of the automatically generated plans and the manual A detection processing unit that calculates the difference between the evaluation result of the modification plan and a proposal process that recommends that it is time to update the weight of the evaluation index in order to minimize the difference, based on the fact that the difference is larger than a predetermined threshold. and an output device that presents to a planner the adjustment of the weight of the evaluation index recommended by the proposal processing unit.

Further, according to the present invention, there is provided a plan creation method in which a manually revised plan obtained by modifying an automatically generated plan by a planner is the final plan, the automatically generated plan and the automatically generated plan being revised by the planner. A plurality of predetermined evaluation indicators are evaluated for each manually revised plan, and the difference between the evaluation result of the automatically generated plan and the evaluation result of the manually revised plan is calculated, and if the difference is larger than a predetermined threshold, the difference is minimized. 1. A plan creation method characterized by presenting to a plan creator that it is time to update the weights of evaluation indicators for the purpose of

According to the present invention, by automatically updating the equipment at an appropriate timing without unnecessarily updating the equipment, it is possible to achieve sustainable and stable production.

1 is a diagram showing an example of the configuration of a planning support device 10. FIG. A diagram showing an example of a plan. The figure which shows the data structure example of plan information. The figure which shows the data structure example of product information. The figure which shows the example of the data structure of an automatically generated plan result. The figure which shows the data structure example of a manual modification plan result. The figure which shows the example of the data structure of a modification log. FIG. 2 is a diagram showing time-series processing of various functions within the central processing unit of FIG. 1; 9 is a flowchart specifically showing various functions shown in FIG. 8; A diagram chronologically summarizing the amount of constraint violations and evaluation index values for each month. A diagram showing the results after correction. The figure which shows the example of a screen which recommends the variation value of the difference between an automatically generated plan and a revised plan.

Embodiments of the present invention will be described below with reference to the drawings. Note that the plans that can be targeted by the planning device are not limited, and can be applied to a wide variety of plans such as factory production, maintenance of social infrastructure, or personnel assignments in various facilities. An example of a planning support device that creates a so-called production plan for manufacturing a product by working with the system will be explained.

In an embodiment of the present invention, a case will be described in which a part of the planning device is configured as a planning support device that is left to the discretion of the planner.

A planning device according to an embodiment of the present invention will be explained using FIG. 1. Note that although the specific configuration for realizing the planning support device 10 will be explained assuming a mainframe, a personal computer, etc., it is also possible to implement it on a cloud.

The planning support device 10 includes the following hardware configuration. That is, a storage device 120 configured with a non-volatile storage device such as an SSD (Solid State Drive) or a magnetic medium such as a hard disk drive, a memory 150 configured with a volatile storage device such as a RAM, and a program 130 held in the storage device 120. A central processing unit (arithmetic unit) 110 that executes the program by reading it into the memory 150, performs overall control of the device itself, and performs various judgments, calculations, and control processes, accepts key inputs and voice inputs from the plan creator M. It includes an input device 160 and an output device 140 such as a display that displays processed data, and these pieces of hardware are communicably connected to each other via a bus.

The central processing unit 110 reads the program 130 stored in the storage device 120 into the memory 150 and executes it, thereby causing an evaluation processing unit 111 that evaluates the automatically generated plan and the manual modification plan, and detects situations that should be dealt with from the evaluation results. A detection processing unit 112 that performs automatic generation planning, a proposal processing unit 113 that requests the plan creator M to adjust the weights of evaluation indicators when the difference in evaluation values becomes large, an adjustment processing unit 114 that adjusts the weights of evaluation indicators, and an automatic generation plan. Each function of the automatic planning processing unit 115 to be created is executed.

With the above configuration, the planning support device 10 according to the embodiment of the present invention evaluates the difference between the automatically generated plan created by the automatic planning processing unit 115 and the manually revised plan manually revised by the planner, and automatically It supports optimization and adjustment of the functions of the planning processing unit 115.

The program 130 may be stored in the storage device 120 or may be introduced into the storage device 120 from another device via a predetermined medium by the planning support device 10 when necessary such as during execution. The medium refers to, for example, a storage medium that is removable from a predetermined interface of the planning support device 10, or a communication medium.

The storage device 120 also stores data necessary to execute these functions, specifically, constraints/objective functions (planning conditions) 121, plan-related information 122, automatically generated plan results 123, and manually revised plans. A result 124, an evaluation result 125, a weight 126 of each evaluation, and a program 130 are stored.

The above-mentioned various data stored in the storage device 120 in FIG. 1 can be classified into preset information, sequentially updated information, and information generated by the present invention. The information generated by the present invention is a constraint/objective function (planning condition) 121 and plan-related information 122, and the information that is updated sequentially is an automatically generated plan result 123 and a manually revised plan result 124. The information generated by the present invention is evaluated. It can be said that they are the result 125 and the weight 126 of each evaluation.

Next, the details of each data will be explained, but as a premise, an example of a plan created by the plan creation support device 10 will be explained first with reference to FIG. In FIG. 2, it is assumed that a production plan is created to produce M products Pi (i=1 to M) by executing N processes Lj (j=1 to N), and for example, a production plan as shown in the Gantt chart 20 is created. Aim to get a plan. Note that the Gantt chart 20 is an example when N=4 and M=3, and also shows the time axis for the process L4.

FIG. 2 also shows planning conditions for the planning support device 10 to create a plan as shown in the Gantt chart 20. Furthermore, when creating a production plan, a set Pi of products and a set Li of processes necessary to produce the products are defined, and the processing time T _ij of each process for each product is given as an explanatory variable.

In the example of FIG. 2, two conditions are provided as constraint conditions. The first is an order constraint, which is a condition that in each process, processes for other products cannot be started until the process for the current product is completed. The second is equipment constraints, which is a condition that in each process, work on multiple products cannot be performed in one process. Furthermore, as a goal of the plan, an objective function f is set that minimizes the time to complete all processes for all products (total processing time). Therefore, in this case, creating a production plan means finding the start time t _ij (decision function) of each process for each product that satisfies the above two constraints and minimizes the objective function f. . Note that the above is just an example, and other constraint conditions such as combination conditions of product types, delivery date conditions, or objective functions such as minimization of production costs may be set. Further, a plurality of objective functions may be set.

The data of the above constraints, objective functions, decision variables, and explanatory variables are stored as constraints/objective functions (planning conditions) 121 in the storage device 120. Next, the data structure of data used by the planning support device 10 to create a plan will be explained.

The plan-related information 122 is a collection of plan-related data collected about past plans. The contents of the plan-related information 122 can be arbitrarily selected by the plan creator M as information that influences the quality of the plan. For example, when a planner modifies a plan, it is desirable to decide what kind of data to accumulate as the plan-related information 122 based on knowledge such as what information is used to determine whether or not modification is necessary. As specific examples of the plan-related information 122, plan information 122a and product information 122b are illustrated here.

Each record of the plan information 122a illustrated in FIG. 3 stores values such as a plan number 2211, a plan execution month 2212, a plan execution date 2213, and a number of tasks 2214. The plan number 2211 is identification information that uniquely identifies a plan, and records with the same plan number represent information explaining the plan with the corresponding plan number. The plan execution month 2212 is the execution month of the plan, the plan execution date 2213 is the plan execution date, and the number of operations 2214 is the total number of operations in the plan (number of processes N x number of products M). If the month or date in which the plan is executed and the overall load of the plan are considered to affect the quality of the plan, these data are collected as information 122a explaining the plan. In addition, data such as identification information that uniquely identifies the planner, operating status of production equipment, and weather may also be included. Furthermore, unless the planner M considers the data to be information that explains the plan, even the data illustrated in FIG. 3 need not be included in the plan information 122a. This also applies to other data described below.

Each record of the product information 122b illustrated in FIG. 4 stores values such as a plan number 2221, a product 2222, a color 2223, and an orderer 2224. The plan number 2221 is identification information that uniquely identifies the plan information, and is the same as the plan number 2211. The product 2222 is identification information that uniquely identifies the product targeted by the plan, the color 2223 is the color of the product, and the orderer 2224 is a code that identifies the orderer who ordered the product. The data collected as the product information 122b is selected as product information that affects the quality of the plan. For example, the color information is included in this example assuming that the plan includes a painting process and the color of the product affects the painting order. In addition, data such as product type and delivery date may also be included.

Next, the automatically generated plan result 123 and the manually modified plan result 124 will be explained. The planner may modify the start time determined by the planning support device 10 based on his or her own knowledge and know-how, as necessary, and execute the revised plan. In such a case, the plans before and after modification are stored as an automatically generated plan result 123 and a manually revised plan result 124, respectively. Therefore, the plan (before modification) stored in the automatically generated plan result 123 has the corresponding plan (after modification) stored in the manually modified plan result 124 .

Here, the automatically generated plan result 123 is generated using the automatic plan generation processing section 115, but a specific method for creating the automatically generated plan result 123 in the automatic plan generation processing section 115 is described in Patent Document 1, for example. Since it is well known, a detailed explanation will be omitted here. The present invention does not limit the automatic generation plan creation method in the automatic plan creation processing section 115.

Each record of the automatically generated plan result 123 illustrated in FIG. 5 stores values such as a plan number 2301, a product 2302, a process 2303, a processing time 2304, and a start time 2305. The plan number 2301 is identification information that uniquely identifies the plan, and is the same as the plan number 2211. The product 2302 is identification information that uniquely identifies the product targeted by the plan, the process 2303 is identification information that uniquely identifies the process targeted by the plan, and the processing time 2304 is the processing time T of the applicable process for the applicable product. _ij , start time 2305 is the start time _tij of the corresponding process for the corresponding product. It will be understood that the Gantt chart shown in FIG. 2 can be created from the information of records having the same plan number. The field values of product 2302, process 2303, and processing time 2304 are values given for creating a plan, and the field value of start time 2305 is a value determined by the plan creation support device 10.

Each record of the manual modification plan result 124 illustrated in FIG. 6 stores values such as a plan number 2401, a work 2402, a process 2403, a processing time 2404, and a start time 2405. The meanings of the items in each record are the same as the items in the automatically generated plan result 123 shown in FIG. By comparing the two, it can be seen that, for example, in the plan number 0601-0800, the start time _t21 of product P2 and process L1 has been revised from 08:10 to 08:20.

Note that it is preferable to further store a modification log for the manual modification plan result 124 in the storage device 120. By storing the modification log and presenting it as appropriate, it can be expected to support the decision of the plan creator. The modification log is a modification log in which actions modified by the plan creator are recorded step by step. Each record of the modification log 125 illustrated in FIG. 7 stores values such as a plan number 2501, a step 2502, a product 2503, a process 2504, a start time before modification 2505, and a start time after modification 2506.

The plan number 2501 is identification information that uniquely identifies the plan, and is the same as the plan number 2211. Step 2502 is the order of actions revised by the plan creator in the plan with the corresponding plan number, starting from number 1 for each plan number. The product 2302 is the identification information that uniquely identifies the product that is the target of modification, the process 2303 is the identification information that uniquely identifies the process that is the target of modification, and the pre-modification start time 2505 is the identification information that uniquely identifies the product that is the target of modification. This is the start time t _ij , that is, the start time determined by the planning support device 10 when creating the plan, and the corrected start time 2506 is the start time t _ij corrected by the plan creator. The modification log 125 is stored as information indicating the importance of each modification, since it is thought that the planner makes important modifications first, such as modifications that have a large impact on other decision variables. be.

The storage contents of the storage device 120 have been described above. Among the various data stored in the storage device 120, the evaluation results 125 and the weights 126 of each evaluation index are information generated by the present invention. The specific details will be explained separately.

Returning to FIG. 1, the planning support device 10 in FIG. 1 uses various functions shown in the central processing unit 110 to execute calculations using various data stored in the storage device 120, and produces intermediate products as a result. They want a final product. The various functions shown in the central processing unit 110 are processed in chronological order as shown in the flow of FIG.

According to FIG. 8, an automatically generated plan 123 and a manual modification plan 124 are inputted to the central processing unit 110, appropriate information is displayed on an output device 140 such as a monitor, and a plan creator M can input appropriate information from the input device 160. By providing the instruction input, the function of the automatic planning processing unit 115 is supported to be optimized.

In the internal processing of the central processing unit 110, the evaluation processing unit 111 first inputs the automatic generation plan 123 and the manual correction plan 124. Here, the automatically generated plan 123 is automatically created by the automatic planning processing unit 115, and the manually revised plan 124 is created by the plan creator M after looking at past driving records and taking into account his own know-how. This is a modification of the automatically generated plan 123, and the plant and the like are finally operated according to the manual modification plan 124. The evaluation processing unit 111 inputs the evaluation results of the automatically generated plan 123 and the manual correction plan 124 for a certain period in the past. In the following explanation, it is assumed that the evaluation results of the monthly automatic generation plan 123 and manual correction plan 124 are input for about half a year.

The evaluation processing unit 111 evaluates two plans, the automatically generated plan 123 generated by the automatic planning processing unit 115 and the manually revised plan 124 manually revised by the plan creator M. When performing the evaluation, the function of checking violation of the constraint C and the function of calculating the evaluation index value E, which are installed in the automatic planning processing unit 115, are used. The evaluation results are stored and accumulated in the storage device 120 as 125.

The detection processing unit 112 presents the evaluation result 125 to the plan creator M through a comparison visualization 130 through an output device 140 such as a monitor. The proposal processing unit 113 compares and evaluates the automatically generated plan 123 and the manually modified plan 124, and when the difference in evaluation values becomes large, the proposal processing unit 113 makes an update recommendation 131 to adjust the weight of the evaluation index and outputs the plan through an output device 140 such as a monitor. Present to creator M. Depending on the content of the presentation, the planner M can appropriately input factors that seem to require adjustment, their magnitude, etc. to the central processing unit 110 via the input device 160, thereby allowing them to be reflected in subsequent processing.

The adjustment processing unit 114 searches for and determines the weight of the evaluation index that reduces the difference in evaluation values when the difference in evaluation values becomes large. This process only searches for weight values, and does not change the weight values that are incorporated into automatic generation during operation. The weight of the evaluation index determined with the approval of the planner M is stored and accumulated in the storage device 120 as 126, and is reflected in the next plan in the automatic planning processing section 115. Note that the determined weight of the evaluation index is appropriately presented to the plan creator M by the comparison visualization 130 through a monitor screen or the like.

FIG. 9 is a flowchart specifically showing the various functions shown in FIG. 8. In the flow of FIG. 9, processing steps S100 to S400 are processings in the evaluation processing unit 111 and detection processing unit 112, processing steps S500 and processing steps S510 are processings in the proposal processing unit 113, and processing steps S520 to S400 are processings in the proposal processing unit 113. Step S540 is a process performed by the adjustment processing section 114, and processing step S600 is a process performed by the detection processing section 112.

First, the processing in the evaluation processing section 111 and the detection processing section 112 will be explained. In the flow of FIG. 9, first, in processing step S100, one or more sets of a manually revised plan drawn up by the plan creator M and an automatically generated plan generated by the automatic planning processing unit 115 are prepared.

Next, in processing step S200, the evaluation processing unit 112 receives the manual correction plan and the automatically generated plan, and calculates the evaluation value of the evaluation index. Below, evaluation indicators and their evaluation values for manually modified plans and automatically generated plans will be explained in detail using examples.

Here, a constraint condition C and an evaluation index value E are assumed as evaluation indices. Constraint C is a matter that should be improved (so to speak, a negative evaluation factor) such as "must not be done, or should not be done if possible", and evaluation index value E is "desirable to last as long or as large as possible". ” are things that should be recommended (so to speak, positive evaluation factors). The evaluation value is a numerical representation of these positive and negative evaluation indicators.

In the example shown below, three cases from constraint C01 to constraint C03 are set for constraint C as shown in the number # and classification column on the left side of FIG. Assume that three cases of evaluation index value E03 are set.

For example, these are as follows. Constraint C01 is a constraint that in each process, a process for another product cannot be started unless the process for the current product is completed, and constraint C02 is a constraint that in each process, a single process cannot perform work on multiple products. The constraint C03 is a constraint that production must be completed by the delivery date set for each product.

Regarding the evaluation index value E, the evaluation index value E01 is to highly evaluate a plan with a small number of appropriate inventory days in terms of the number of inventory days, and the evaluation index value E02 is to evaluate a plan with a small number of appropriate inventory days in terms of inventory days, and an evaluation index value E02 is to evaluate a plan in which the same product is continuously produced with respect to the number of consecutive productions of the same product. The evaluation index value E03 is such that a plan in which the mold change time is as short as possible in one day in terms of the ratio of single item production/day is highly evaluated.

Looking at the three cases from constraint condition C01 to constraint condition C03, the smaller the value of each constraint condition, the better the plan. Moreover, looking at the three cases of evaluation index E01 to evaluation index E03, the smaller the value of any evaluation index, the better the plan.

Returning to FIG. 9, in processing step S200, as a result of the processing of the evaluation processing unit 111, the automatically generated plan 123 generated by the automatic planning processing unit 115 and the manually revised plan 124 that was manually revised by the plan creator M. Assume that the plan evaluation shown in FIG. 10 is obtained as a result of evaluating each of the two plans.

FIG. 10 shows a chronological summary of the amount of violations of a plurality of constraint conditions C and the evaluation results of a plurality of evaluation index values E for each month. The symbol ● is an automatically generated plan, and the symbol ▲ is a manually modified plan. In FIG. 10, the amount of violation of multiple constraints C and the evaluation results of multiple evaluation index values E are compared in chronological order for each month of the four months from January to April shown on the horizontal axis. The amount of violation on the vertical axis indicates that there was no violation when it is zero. This evaluation result is stored and accumulated in the evaluation result 125 of the storage device 120. Note that hereinafter, the amount of violation of the constraint C and the evaluation results of the plurality of evaluation index values E may be referred to as an evaluation value.

Note that the trends in FIG. 10 presented by the detection processing unit 112 in FIG. 9 represent various events. For example, in April, it was detected that the difference in evaluation values between the automatically generated plan shown by the dark line and the manual modification plan shown by the light line became large, and as shown in Figure 10, The results of a test plan generated by adjusting the weights of evaluation indicators may also be shown.

Testing here means that when the difference in evaluation values becomes large, a plan is generated by applying the weight of the adjusted evaluation index to the automatically generated plan, the newly generated plan is evaluated, and the automatically generated plan before adjustment is The plan is then displayed to the plan creator M side by side with the evaluation results of the plan and the manually revised plan.

Furthermore, the following can be seen from the changes in the amount of violations of the constraint C and the fluctuations in the evaluation index E illustrated in FIG. In this example, the automatically generated plan generally produces a better plan than the manually modified plan. The amount of violations of the constraints C01, C02, and C03 is greater in the manually modified plan than in the automatically generated plan, indicating that violations of these constraints are allowed to some extent. In particular, the amount of violations of the manual modification plan for constraint C03 is on the rise, indicating that this constraint is not important. The evaluation indicators E01, E02, and E03 are similar for both the automatically generated plan and the manually corrected plan.

At this time, the detection processing unit 112 performs comparison visualization 130 on the output device 140, and arranges the evaluation results of the automatic generation plan 123 and manual correction plan 124 stored in the storage device 120 in chronological order, for example, as shown in FIG. It is best to display it in a format like this.

These events detected by the processing of the detection processing unit 112 include events that should be reflected in the automatic plan in the automatic planning processing unit 115, so the proposal processing unit 113 includes them in the automatic generation plan. Comparatively evaluate the manually revised plan, and when the difference in the evaluation values becomes large, the plan creator M is presented with an adjustment of the weight of the evaluation index. This process is the process of process step S300 in the flow of FIG.

In the evaluation process of processing step S300, the difference between the evaluation results of the automatically generated plan and the manually modified plan is calculated and stored and accumulated in the storage device 120 together with the evaluation result 125. Here, regarding the difference in evaluation results, for example, the difference di in the evaluation results of the i-th plan is calculated using equation (1).

In equation (1), pj is the value of the j-th evaluation index in the automatically generated plan, and pj' is the value of the j-th evaluation index in the manually modified plan. When comparing the quality of a plan, cosine similarity was used instead of Euclidean distance because the balance of priorities between different evaluation indicators has a greater influence than the size of the evaluation index value.

In the evaluation process of processing step S400, the detection processing unit 112 uses equation (2) to calculate a variation value of the difference between the evaluation results of the automatically generated plan and the manually corrected plan. Specifically, the variation value α is calculated from the difference between the evaluation values of the plan set calculated in the preparation stage and the difference between the evaluation results of the latest plan using equation (2) in accordance with Hotelling's theory. Here, d0 and σ0 are the average value and standard deviation of the difference between the evaluation values of the plan set calculated in the preparation stage. Further, dn is the difference between the evaluation results of the latest plans. The above is the processing in the evaluation processing section 111 and the detection processing section 112.

Next, in the process in the proposal processing unit 113, first, in processing step S500, the variation value α is compared with a threshold value k, and when the variation value α is greater than or equal to the threshold value, the process proceeds to processing step S510, and when it is less than the threshold value, the process proceeds to processing step S600. Since the variation value α follows a chi-square distribution, a threshold value k=2.71 is set at which a change is recognized with a probability of 90% or more. In processing step S510, the proposal section 113 recommends via the output device 140 that it is time to update.

Next, in the process in the adjustment processing unit 113, first in processing step S520, the weight of the evaluation index is adjusted so that it approaches the evaluation index of the modified plan. Specifically, the weights of the evaluation indicators are adjusted by solving the following inverse optimization problem. The obtained weight of the evaluation index is stored and accumulated in the storage device 120 as 126, and is reflected in the next processing by the automatic plan creation unit 115.

The calculation formula (objective function) f(x) of the planned score in automatic generation can be expressed by equation (3) from the planning value x, the calculation formula fi(x) of the i-th evaluation index, and the weight wi.

Also, there is an inverse problem E(w) for calculating the weight w of an unknown evaluation index from the planned value x ^* obtained by the manual modification plan, and it can be expressed by equation (4).

Note that the inverse problem E(w) is solved using a local search method with the initial solution as the weight of the currently set evaluation index. The planned value x is determined by automatic planning in which the weight w of the evaluation index is set.

Next, in processing step S530, the automatic planning processing unit 115 solves the plan optimization problem by reflecting the adjusted weights and outputs it as a test plan. The test plan process here generates a plan by applying the adjusted evaluation index weights to the automatic plan. Then, an evaluation index value is calculated for the test plan by using a violation checking function for constraint C and a function for calculating evaluation index value E, both of which are installed in the automatic planning processing unit 115.

In processing step S540, the automatic planning processing unit uses equation (1) to calculate the difference between the test plan generated in processing step S530 and the manually revised plan received in processing step S200.

As for the calculated test results, the detection processing unit 112 displays each evaluation index value of the automatically generated plan and the manual correction plan in chronological order, and also displays the fluctuation value of the difference between the automatically generated plan and the manual correction plan in chronological order. Display in . Upon receiving this display result, the planner M can confirm the test result. Note that the adjusted weights may not be used for test planning, but may be directly reflected in the device for self-running generation planning.

In addition, in processing step S530, the new plan (for example, a test plan) obtained by solving the plan optimization problem reflecting the adjusted weights can be referred to as a second automatically generated plan to distinguish it from the original automatic plan. Therefore, the difference between the evaluation result of the second automatically generated plan and the evaluation result of the manually modified plan can be called a second difference value.

FIG. 11 is a diagram showing the results after correction. The results of weight adjustment reflecting the evaluation values from January to the current month in the evaluation values of the automatically generated plan and manual modification plan in Figure 10 are reflected in the evaluation values from April to June. The test results are shown as dotted lines.

In this example, the weight of each evaluation index (constraint C and evaluation index value E) of the automatically generated plan and manually modified plan before weight adjustment, and therefore when the evaluation values shown in Figure 10 are obtained, is expressed as "1.0". ing. This represents that the weights of the constraint conditions C01 and C03 and the evaluation index values E01 and E03 are reduced, and the weight of the evaluation index value E02 is increased.

In this case, the April and May test results are closer to the manually modified plan than the automatically generated plan, indicating an improvement. In particular, the test results for April's constraint C03 and evaluation index value E03 indicate that the manual correction plan is approaching. However, the test results in June showed that the manual correction plan had moved away and was getting worse.

FIG. 12 shows an example of a screen that recommends the variation value of the difference between the automatically generated plan and the revised plan. An enlarged display of the individual project with constraint C and evaluation index value E in Figure 10 shows that the degree of deviation between the automatically generated plan before adjustment and the manually revised plan for the previous three months has been increasing. , notifies plan creator M that the degree of deviation has become large as of April and recommends weight adjustment, and as a result, the degree of deviation between the automatically generated plan after adjustment and the manually revised plan for the last three months is The test results show that improvements have been made.

To be more specific, the vertical axis in Figure 12 represents the difference between the manually modified plan and the automatically generated plan, or the difference between the test plan and the range is from 0 to 1, with 0 indicating no difference and matching. means that In addition, by converting the variation value to the difference from the manual correction plan using equation (2), we can calculate the difference that can be expected to change with a 90% probability (=0.412, variation value = 2.71) and the 99% Colors are used to indicate the difference that can be expected to change with probability (= 0.500, variation value = 6.63).

In April, the difference in evaluation value between the automatically generated plan and the manually modified plan was 0.415 (variation value = 2.80), which exceeded the threshold, although slightly, so April, May, and June were used as evaluation indicators. The results of the test plan generated by adjusting the weights are also shown. The difference in evaluation value between the test plan generated by adjusting the weights and the manually revised plan as of April was 0.163 (variation value = 1.00), which was less than the threshold and a significant improvement.

Although it was difficult to understand the overall difference from the manual correction plan from the changes in the amount of violation of constraint C and the fluctuation of evaluation index value E in FIG. 10, it became clearer in FIG. It can be seen that the difference from the manually revised plan is particularly large in April. This is considered to be influenced by the constraint C03 and the evaluation index E03. As a result of generating a test plan by adjusting the weights of the evaluation indicators according to the manually revised plan as of April, we were able to create a plan with smaller differences than the original automatically generated plan in April and May.

According to the present invention described above, the function of the automatic planning processing unit 115 is optimized by evaluating the difference between the automatically generated plan created by the automatic planning processing unit 115 and the manually revised plan manually revised by the planner. It can help to make adjustments possible. This makes it possible to achieve sustainable and stable production by automatically updating the equipment at an appropriate time without updating the equipment unnecessarily.

In the above explanation, an example was shown in which the invention is constituted as a support device that presents information to the plan creator at an appropriate timing and proceeds with the process in response to instructions from the plan creator. It is also possible to use an automatic plan creation device that automatically proceeds with the process.

10: Planning support device 110: Central processing unit 111: Evaluation processing unit 112: Detection processing unit 113: Proposal processing unit 114: Adjustment processing unit 115: Automatic planning processing unit 121: Constraint conditions/objective function (planning conditions)
122: Plan related information 123: Automatically generated plan result 124: Manually modified plan result 125: Evaluation result 126: Evaluation weight 120: Storage device 130: Program 150: Memory 140: Output device 160: Input device

Claims (12)

A plan creation device that uses a manually modified plan obtained by modifying an automatically generated plan by a planner as a final plan,
a storage device that stores an automatically generated plan created by an automatic planning processing unit and a manually modified plan that is modified by a plan creator;
an evaluation processing unit that evaluates a plurality of predetermined evaluation indicators for the automatically generated plan and the manual correction plan; and a detection processing unit that calculates a difference between the evaluation result of the automatically generated plan and the evaluation result of the manual correction plan. , a central processing unit comprising a proposal processing unit that recommends that it is time to update the weight of the evaluation index in order to minimize the difference when the difference is larger than a predetermined threshold;
A plan creation device comprising: an output device that presents to a plan creator adjustment of the weight of the evaluation index recommended by the proposal processing unit. The planning device according to claim 1,
The plan creation device is characterized in that the evaluation processing unit obtains, as the evaluation result, an evaluation value obtained by quantifying a predetermined evaluation index for each of the automatically generated plan and the manually modified plan. The planning device according to claim 2,
A planning device characterized in that the evaluation index includes a plurality of sets of constraint conditions that are items to be improved and evaluation indicators that are items to be recommended. The planning device according to claim 1,
The plan creation device, wherein the central processing unit includes an adjustment processing unit that adjusts the weight of the evaluation index of the automatically generated plan so that it approaches the evaluation index of the manually modified plan. The planning device according to claim 4,
The automatic planning processing unit creates a second automatically generated plan based on the adjusted weights, and generates a second automatically generated plan that is the difference between the evaluation result of the second automatically generated plan and the evaluation result of the manually modified plan. A plan creation device characterized by calculating two difference values and presenting at least the second difference values in time series on the output device. The planning device according to claim 1,
A plan creation device, wherein the output device presents evaluation results of the automatically generated plan and evaluation results of the manually modified plan in chronological order. The planning device according to claim 1,
Furthermore, it includes an input device into which an instruction regarding adjustment according to the content presented by the output device is input,
The plan creation device is characterized in that the central processing unit obtains instructions input to the input device and proceeds with processing based on the instructions. A plan creation method in which a manually modified plan obtained by modifying an automatically generated plan by a planner is the final plan,
A plurality of predetermined evaluation indicators are evaluated for the automatically generated plan and the manually revised plan that is modified by the plan creator, and the difference between the evaluation result of the automatically generated plan and the manually revised plan is calculated. is calculated, and when the difference is larger than a predetermined threshold, the plan creator is informed that it is time to update the weights of evaluation indicators to minimize the difference. . 9. The planning method according to claim 8,
A plan creation method characterized in that, for each of the automatically generated plan and the manually modified plan, an evaluation value obtained by quantifying a predetermined evaluation index is obtained as the evaluation result. The plan creation method according to claim 9,
The plan creation method is characterized in that the evaluation index includes a plurality of sets of constraints, which are items to be improved, and evaluation indicators, which are items to be recommended. 9. The planning method according to claim 8,
A plan creation method characterized by adjusting the weight of the evaluation index of the automatically generated plan so as to approach the evaluation index of the manually modified plan. 12. The planning method according to claim 11,
Create a second automatically generated plan based on the adjusted weights, calculate a second difference value that is the difference between the evaluation result of the second automatically generated plan and the manually modified plan, and then A planning method characterized by presenting.

Images