Plan generation method and plan generation system

Abstract

Generate a plan that reflects the intentions of the planner. [Solution] A plan generation method executed by a plan generation system, the plan generation system comprising a calculation unit and a storage unit, the storage unit holding a base plan which is a plan for a business, resource information relating to resources allocated to the business in the base plan, and a learning model which outputs a recommended plan including changes recommended to the base plan and an estimation reason which is an estimation of the reasons for the changes in the recommended plan, the plan generation method comprising: a step in which the calculation unit generates a recommended plan based on the base plan, resource information and the learning model, and a step in which, when a revised plan which is a modified recommended plan and the reasons for the revision thereof are input, the calculation unit retrains the learning model using the revised plan and the reasons for the revision as the correct data for the recommended plan and the estimation reason, respectively.

Description

Technical Field

[0001] The present invention relates to tasks such as formulating procurement plans and production plans in the supply chain.

Background Art

[0002] In modern manufacturing and distribution industries, tasks such as formulating production plans require plans that reflect the ever-changing situations, such as the work schedules of workers. For example, a production plan determines the production start date and production volume of a product in consideration of demand forecasts, inventory outlooks, the production capacity of factories, etc.

[0003] These various plans are required to be suitable for the situation. In the case of a production plan, by formulating a plan according to the work schedule of workers, etc., the efficiency of the production line can be maximized, and reduction of the defect rate and improvement of throughput can be achieved.

[0004] However, in the manufacturing site, information such as the work schedules of workers changes daily, so the planners have to grasp and modify them each time, which requires man-hours. For example, consider a use case where it is desired to change the production plan of a certain product to be produced on a production line with many skilled workers. In this case, the planner modifies the production plan based on the work schedules of the skilled workers. As background art for reducing such modification man-hours, there is Japanese Unexamined Patent Application Publication No. 2019-46287 (Patent Document 1). Patent Document 1 discloses a method for generating a plan that has learned the modification history.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] Patent Document 1 describes "acquiring information on revisions to the plan from a predetermined device, and identifying the relationship between the plan and the revised plan, and the values ​​related to predetermined events for plan evaluation, using a predetermined algorithm." However, a problem with Patent Document 1 is that, because it generates a plan by learning only the revision history, it learns in a way that does not align with the intentions of the planner, requiring revisions each time until the plan meets the planner's requirements, and thus failing to reduce man-hours. For example, when a product's production line is changed, there may be reasons for the change such as "to produce the product on a production line with many skilled workers." However, the current method, which learns only the revision history, cannot capture such reasons for change and learns only the information that "the production line for the product in question has been changed." As a result, when the assignment of skilled workers is moved during the next planning stage, it may not be possible to change the production plan to match the skilled workers' work schedules.

[0007] As described above, the result is that revisions are required for the same reasons, and the workload for the planners in making those revisions is not reduced.

[0008] Therefore, by incorporating not only the difference between the pre- and post-revision plans but also the reasons for the revisions into the learning process, the plan that the planner desires can be generated. [Means for solving the problem]

[0009] To solve at least one of the above problems, the present invention provides a plan generation method to be executed by a plan generation system, wherein the plan generation system comprises a calculation unit and a storage unit, the storage unit holds a base plan which is a plan for a business, resource information relating to resources allocated to the business in the base plan, a recommended plan which includes changes recommended to the base plan, and an estimated reason which is an estimation of the reasons for the changes in the recommended plan, and the plan generation method is characterized in that the calculation unit generates the recommended plan based on the base plan, the resource information, and the learning model, and when a modified plan which is a modified version of the recommended plan and the reasons for the modification are input, the calculation unit retrains the learning model using the modified plan and the reasons for the modification as the correct data for the recommended plan and the estimated reason, respectively. [Effects of the Invention]

[0010] According to one aspect of the present invention, by generating a plan that reflects the intentions of the planner, it is possible to achieve the output of a plan desired by the planner, thereby reducing the man-hours required for planning work.

[0011] Other issues, configurations, and effects not mentioned above will be clarified by the following description of the embodiments. [Brief explanation of the drawing]

[0012] [Figure 1] This block diagram shows the functional configuration of the automated plan generation system in this embodiment. [Figure 2] This is a flowchart illustrating the operation of the calculation unit of the automated plan generation system in this embodiment. [Figure 3] This is an explanatory diagram of the supply chain master information held by the automated plan generation system of this embodiment. [Figure 4] This is an explanatory diagram of the personnel information held by the automated plan generation system of this embodiment. [Figure 5] This is an explanatory diagram of the planning information held by the automated planning generation system of this embodiment. [Figure 6]It is an explanatory diagram of the correction information held by the plan automatic generation system of this embodiment. [Figure 7] It is an explanatory diagram of the reason score information held by the plan automatic generation system of this embodiment. [Figure 8] It is an explanatory diagram of the reason item relationship information held by the plan automatic generation system of this embodiment. [Figure 9] It is an explanatory diagram of the information displayed by the display unit of the plan automatic generation system of this embodiment. [Figure 10] It is a flowchart for explaining the operation of the generation unit of the plan automatic generation system of this embodiment. [Figure 11] It is a flowchart for explaining the operation of the learning unit of the plan automatic generation system of this embodiment. [Figure 12] It is a flowchart for explaining the operation of the correction difference calculation unit of the plan automatic generation system of this embodiment.

Mode for Carrying Out the Invention

[0013] Hereinafter, embodiments will be described, but the scope of the present invention is not limited to the embodiments.

[0014] Hereinafter, embodiments of the present invention will be described based on the drawings.

[0015] FIG. 1 is a block diagram showing the functional configuration of the plan automatic generation system 100 of this embodiment.

[0016] Focusing on functions, the plan automatic generation system 100 has a storage unit 110, a display unit 120, an arithmetic unit 130, and a communication unit 140. Necessary information in the storage unit 110, display unit 120, arithmetic unit 130, and communication unit 140 can be obtained via a communication network 200 or the like from, for example, a system such as ERP (Enterprise Resource Planning), or a database storing data conforming thereto, or simply a storage device.

[0017] The memory unit 110 includes an input information memory unit 111, a plan memory unit 112, a correction information memory unit 113, a reason score memory unit 114, and a reason item relationship memory unit 115. The memory unit 110 is composed of, for example, a storage device such as a hard disk drive or a flash memory device.

[0018] The display unit 120 includes a plan display unit 121, a corrected plan reception unit 122, a presumed reason display unit 123, and a corrected reason reception unit 124.

[0019] The arithmetic unit 130 includes a generation unit 131, a learning unit 132, and a correction difference calculation unit 133.

[0020] The communication unit 140 is an interface device for communicating with the user terminal device 300, which is an external device, via the communication network 200. The communication unit 140 may use either wired communication or wireless communication.

[0021] FIG. 2 is a flowchart for explaining the operation of the arithmetic unit 130 of the plan automatic generation system 100 of the present embodiment. Hereinafter, each step in FIG. 2 will be described.

[0022] (FIG. 2: Step S101: Obtain a reference plan and input information) Step S101 in FIG. 2 is a process for obtaining the reference plan 1211, the supply chain master information 11110, and the human resource information 11120 used for plan generation performed by the generation unit 131. The reference plan 1211, the supply chain master information 11110, and the human resource information 111 – 20 are input via the communication network 200.

[0023] FIG. 5 is an explanatory diagram of the plan information 1120 held by the plan automatic generation system 100 of the present embodiment.

[0024] The plan information 1120 holds, in addition to the reference plan 1211 obtained in step S101, a recommended plan 1212 and a corrected plan 1221, which will be described later. Here, the reference plan 1211 will be described.

[0025] The base plan 1211 is a plan that is input into the automatic plan generation system 100 from another planning system, among the plans held as planning information 1120. As shown in Figure 5, the planning information 1120 holds the plan ID 1121, planning date 1122, scheduled date 1123, item ID 1124, planned quantity 1125, actual value 1126, target plan 1127, plan type 1128, and attribute information 1129.

[0026] Plan ID 1121 is the primary key that uniquely identifies the plan. Plan Date 1122 is the date the plan was created. Scheduled Date 1123 is the date the plan will be executed. Item ID 1124 identifies the item of goods that are the subject of the plan. Planned Quantity 1125 is the planned quantity (for example, in a production plan, the planned production quantity for the item). Actual Value 1126 is the actual value relative to the Planned Quantity 1125 (for example, in a production plan, the quantity of goods actually produced according to the plan).

[0027] The target plan 1127 represents the subject of the plan, such as a production plan, procurement plan, or sales plan. The plan type 1128 is used to input whether the plan is a base plan 1211, a recommended plan 1212 (described later), or a modified plan 1221. The attribute information 1129 is used to input information about the target of the plan, such as the line name for a production plan, the supplier name for a procurement plan, or the buyer name for a sales plan. The base plan 1211 is stored in the plan storage unit 112. In this case, the plan type 1128 is entered as "Base Plan".

[0028] The supply chain master information 11110 in Figure 3 and the personnel information 11120 in Figure 4 are information used to generate a recommendation plan 1212 and an estimated reason score 1231 from a baseline plan 1211 using this automated plan generation system 100. The supply chain master information 11110 and personnel information 11120 are stored in the input information storage unit 111.

[0029] Figure 3 is an explanatory diagram of the supply chain master information 11110 held by the automated plan generation system 100 of this embodiment.

[0030] The supply chain master information 11110 holds, for example, location ID 11111, location name 11112, item ID 11113, item name 11114, source 11115, and destination 11116. This information can be obtained from systems such as ERP.

[0031] Location ID 11111 and location name 11112 identify the locations of businesses that make up the supply chain. Item ID 11113 and item name 11114 identify the items of goods handled at each location. Origin 11115 and destination 11116 indicate the origin and destination of goods when the business is a carrier.

[0032] Figure 4 is an explanatory diagram of the personnel information 11120 held by the automated plan generation system 100 of this embodiment.

[0033] Personnel information 11120 holds, for example, employee ID 11121, years of service 11122, qualification A held 11123, evaluation 11124, skill level category 11125, assigned duties 11126, and overtime hours for the current month 11127.

[0034] Information such as employee ID 11121, length of service 11122, qualification A possession 11123, and evaluation 11124 can be obtained from a time and attendance management system (not shown in the diagram). Employee ID 11121 identifies personnel registered in personnel information 11120 (for example, employees that make up each business). Length of service 11122 indicates the length of service for each employee. Qualification A possession 11123 indicates whether each employee possesses a certain qualification (in this example, "Qualification A"). In this example, only the possession of one qualification is shown, but personnel information 11120 may also include information indicating the possession of multiple qualifications. Evaluation 11124 indicates the evaluation of each employee. For example, evaluation 11124 is an evaluation value by a user of this system, and may also be an evaluation value regarding each employee's overall ability or work attitude from another person (for example, each employee's supervisor).

[0035] Skill level classification 11125 is an indicator that shows whether each employee is a beginner or an expert. Skill level classification 11125 may be obtained from information such as years of service 11122, qualification A possession 11123, and evaluation 11124, or it may be obtained from an external human resource management system (not shown in the diagram). For example, suppose that years of service 11122 is entered as "1" if it is 5 years or more, and "0" if it is less than 5 years, qualification A possession 11123 is entered as "1" and not possessed as "0", and evaluation 11124 is entered as a real number in the range of "0" to "1".

[0036] An employee is considered skilled if the average of each element (years of service 11122, qualification A held 11123, evaluation 11124) exceeds "0.5", and their skill level 11125 is set to "1". Conversely, if the average is less than "0.5", the employee is considered a beginner, and their skill level 11125 is set to "0". For example, an employee has 8 years and 2 months of service, holds qualification A, and has an evaluation of "0.7". Considering that the year of service is 5 years or more ("1"), the qualification A is held ("1"), and the evaluation is "0.9", the skill level 11125 is calculated by taking the average of these: (1+1+0.7) / 3 = 0.9, which is 0.5 or higher, so this employee's skill level 11125 can be determined to be "1".

[0037] The "Assignment 11126" field indicates the scope of work each employee is responsible for. For example, in the case of production planning, "Assignment 11126" might indicate the employee's assigned production line; in the case of procurement planning, it might indicate the employee's assigned supplier; and in the case of sales planning, it might indicate the employee's assigned buyer. "Assignment 11126" can also be retrieved from systems such as the personnel master data.

[0038] The "Current Month Overtime Hours 11127" indicates the total overtime hours for each employee during the current month. For example, "Current Month Overtime Hours 11127" represents the total overtime hours from the beginning of the month to the present time, including the current time.

[0039] (Figure 2: Step S102: Generate recommendation plan and estimated reason score) Step S102 in Figure 2 is a process performed by the generation unit 131. The generation unit 131 generates a recommendation plan 1212 and an estimated reason score 1231 from the baseline plan 1211, supply chain master information 11110, and personnel information 11120 acquired in step S101. The specific generation flow of the recommendation plan 1212 and the estimated reason score 1231 is shown in Figure 10.

[0040] Figure 10 is a flowchart illustrating the operation of the generation unit 131 of the automated plan generation system 100 in this embodiment. The steps in Figure 10 will be described below.

[0041] (Figure 10: Step S1311: Load baseline plan and input information) In step S1311 of Figure 10, the generation unit 131 acquires the base plan 1211, supply chain master information 11110, and personnel information 11120, etc., from the plan storage unit 112 and the input information storage unit 111.

[0042] The base plan 1211 is obtained from the plan information 1120 in Figure 5 where the plan type 1128 is "base plan," by referring to the plan date 1122 and the target plan 1127. For example, in the case of a production plan, a plan with the execution date 1122 of this system and the target plan 1127 being the production plan is obtained from the plan storage unit 112.

[0043] Furthermore, in the case of supply chain master information 11110 in Figure 3, information filtered using location ID 11111 and item ID 11113 is read from the input information storage unit 111, and in the case of personnel information 11120 in Figure 4, information filtered using employee ID 11121 is read from the input information storage unit 111.

[0044] (Figure 10: Step S1312: Using the model generated in the learning unit, a recommendation plan and estimated reason score are generated from the baseline plan and input information.) Step S1312 in Figure 10 is a process performed by the generation unit 131. Based on the baseline plan 1211, supply chain master information 11110, and personnel information 11120 received in step S1311, and the learning model generated by the learning unit 132, the generation unit 131 generates a recommendation plan 1212 and an estimated reason score 1231.

[0045] Recommendation plan 1212 is a modified version of the base plan 1211 based on supply chain master information 11110 and personnel information 11120. Like the base plan 1211, recommendation plan 1212 has the information shown in Figure 5. Among the plans included in the plan information 1120, the plan type 1128 for the recommendation plan is entered as "Recommendation Plan". The following explanation will be given in conjunction with each part of the display unit 120 in Figure 9.

[0046] Figure 9 is an explanatory diagram of the information displayed by the display unit 120 of the automatic plan generation system 100 in this embodiment.

[0047] The estimated reason score 1231 displayed in the estimated reason display unit 123 of Figure 9 is information indicating the reason for the change from the base plan 1211 to the recommended plan 1212. For example, it may include at least one of the following: information describing the reason itself, and a score representing the importance of the reason. The estimated reason score 1231 is stored in the reason score storage unit 114 as reason score information 1140. Similarly, the modified reason score 1241, which will be described later, is also stored as reason score information 1140. Now, the reason score information 1140 will be explained with reference to Figure 7.

[0048] Figure 7 is an explanatory diagram of the reason score information 1140 held by the automated plan generation system 100 of this embodiment.

[0049] The reason score information 1140 includes, for example, the plan ID 1141, item 1142, reason score 1143, and reason score type 1144. The plan ID 1141 identifies the generated plan (e.g., a recommended plan or a revised plan). Item 1142 is information indicating the reason for the plan change, such as considering proficiency, reducing the number of setup changes, or leveling out overtime hours. The reason score 1143 is a score representing the importance of item 1142. The reason score type 1144 indicates whether the reason score is an estimated reason score 1231 or a revised reason score 1241.

[0050] In other words, among the reason score information 1140, information in which the plan ID 1141 is the identification information of the recommended plan 1212 displayed on the plan display unit 121 shown in Figure 9, and the reason score type 1144 is "estimated reason score", is displayed on the estimated reason display unit 123 as the estimated reason score 1231.

[0051] On the other hand, the value of the correction reason score 1241 entered into the correction plan reception unit 122 shown in Figure 9, or the value of the estimated reason score 1231 corrected thereafter, is stored as information in the reason score information 1140 where the plan ID 1141 is the identification information of the correction plan 1221 displayed in the correction plan reception unit 122 shown in Figure 9, and the reason score type 1144 is "correction reason score". The correction reason score 1241 is, as will be described in detail later, the reason score 1143 of the items 1142 that the user (for example, the planner) considered important when correcting the recommended plan 1212 generated by the automatic plan generation system 100.

[0052] The estimated reason score 1231 is generated based on the revised reason score 1241. An example of the generation method is shown below.

[0053] First, extract only the modified reason scores from reason score type 1144. Next, calculate the average of the reason scores 1143 for each item listed in item 1142. For example, if we extract only the records where item 1142 contains "overtime leveling," let's assume there are 2 records. Next, assuming the reason scores 1143 for the two records are 3 and 5, the average is 4, which we will use as the importance of the estimated reason score for overtime leveling. Similarly, calculate the importance of each item, such as the importance of the estimated reason score for considering proficiency, the importance of the estimated reason score for reducing the number of setup changes, etc.

[0054] The method for generating recommendation plan 1212 is shown below.

[0055] The generation unit 131 selects item 1142, which has the highest importance in the estimated reason score 1231, modifies the base plan 1211, and outputs the recommended plan 1212. For example, suppose the item "Overtime leveling" has the highest importance. In Figure 9, the base plan 1211 shows that the overtime for line A is -0.5 hours on day 1 and 2.0 hours on day 2. In this case, the automatic plan generation system 100 generates the recommended plan 1212 by moving 25 units of item B from the 75 units produced on day 2 to day 1 so that the overtime for day 1 and day 2 are both 1.0 hours. As a result, the number of setup changes on day 1 increases by 1, but the overtime for both day 1 and day 2 becomes 1.0 hours, generating a recommended plan 1212 that prioritizes overtime leveling over reducing the number of setup changes.

[0056] (Figure 10: Step S1313: Recommendation plan and estimated reason score are stored in the plan storage unit and reason score storage unit.) Step S1313 in Figure 10 is a process performed by the generation unit 131. The plan storage unit 112 stores the recommendation plan 1212 generated in step S1312 as plan information 1120. At this time, the plan type 1128 is entered as "recommendation plan". The reason score storage unit 114 assigns a new ID to the recommendation plan generated in step S1312 and stores that ID in plan ID 1141, stores the generated estimated reason score item in item 1142, and stores the reason score in reason score 1143. At this time, the reason score type 1144 is entered as "estimated reason score".

[0057] (Figure 2: Step S103: Display the baseline plan and recommended plan to the user) Step S103 in Figure 2 is a process performed by the plan display unit 121. The plan display unit 121 displays the base plan 1211 acquired in step S101 and the recommended plan 1212 generated in step S102 to the user.

[0058] To display the base plan 1211, plans whose plan type 1128 is the base plan are extracted from the plan information 1120 in the plan storage unit 112. Furthermore, filtering is performed using the plan date 1122, scheduled date 1123, and target plan 1127 to identify the plans to be displayed. Figure 9 shows the case where the target plan 1127 is a production plan.

[0059] Line information is obtained from attribute information 1129. Product name information is obtained from item name 11114 of the record where item ID 1124 in planning information 1120 matches item ID 11113 in supply chain master information 11110. The daily production quantity shown in the base plan 1211 can be obtained from the planned quantity 1125 of the record where the scheduled date 1123 in planning information 1120, shown in Figure 5, corresponds to each day in the base plan 1211.

[0060] The above example assumes that the target plan 1127 is a production plan. However, if it is a sales plan, the system searches for and displays information such as the location name 11112 of records where the attribute information 1129 of the plan information 1120 (Figure 5) matches the location ID 11111 in the supply chain master information 11110 (Figure 3). Similarly, if it is a transportation plan, the system searches for and displays information on the source 11115 and destination 11116 in the supply chain master information 11110 (Figure 3).

[0061] Recommendation Plan 1212 extracts and displays plans from Plan Information 1120 where Plan Type 1128 is "Recommendation Plan". The extraction method is the same as in the case of Base Plan 1211 described above.

[0062] (Figure 2: Step S104: Display the estimated reason score to the user) Step S104 in Figure 2 is a process performed by the estimated reason display unit 123. The estimated reason display unit 123 displays the estimated reason score generated in step S102 to the user.

[0063] As shown in Figure 9, the estimated reason score 1231 of the estimated reason display unit 123 of the display unit 120 is displayed to the user. The estimated reason items are items 1142 of the reason score information 1140 (Figure 7), such as overtime hour leveling, skill level consideration, and reduction of setup changes, and are shown to the user. The importance is displayed as the reason score 1143 of the reason score information 1140 (Figure 7). In Figure 9, the importance is displayed using numerical radio buttons, but this can be replaced with a slider display or a display using natural language, etc.

[0064] (Figure 2: Step S105: The correction plan is received from the user by the correction plan reception unit and stored in the plan storage unit.) Step S105 in Figure 2 is a process performed by the revision plan reception unit 122 and the plan storage unit 112. The revision plan reception unit 122 receives a revision plan 1221 from the user who has confirmed the base plan 1211 and recommended plan 1212 displayed in step S103. Specifically, the revision plan reception unit 122 of the display unit 120 in Figure 9 receives the revision plan 1221 from the user. Furthermore, the received revision plan 1221 is stored in the plan storage unit 112 as plan information 1120.

[0065] The revised plan 1221 holds, for example, a plan ID 1121, a plan date 1122, a scheduled date 1123, an item ID 1124, a planned quantity 1125, an actual value 1126, a target plan 1127, a plan type 1128, and attribute information 1129, as shown in Figure 5. For the plan ID 1121, enter a value that is a new ID that does not overlap with other records. For the plan date 1122, enter the date the revised plan 1221 was received. For the scheduled date 1123, enter the target date for the revised plan 1221. For the plan type 1128, enter "Revised Plan". For the attribute information 1129, enter the target information for the plan, such as the line name for a production plan, the supplier name for a procurement plan, or the buyer name for a sales plan.

[0066] Furthermore, the correction information storage unit 113 stores the correction information 1130.

[0067] Figure 6 is an explanatory diagram of the correction information 1130 held by the automatic plan generation system 100 of this embodiment.

[0068] As shown in Figure 6, the correction information storage unit 113 of the automatic plan generation system 100 stores correction information 1130 consisting of correction ID 1131, correction date and time 1132, pre-change plan ID 1133, and post-change plan ID 1134. A new ID that uniquely identifies the correction information 1130 is assigned and entered in correction ID 1131. The date and time of assignment is recorded in correction date and time 1132. The plan ID of the recommended plan 1212 is extracted from plan ID 1121 in the plan storage unit 112 and entered in pre-change plan ID 1133. The plan ID of the corrected plan 1221 is extracted from plan ID 1121 in the plan storage unit 112 and entered in post-change plan ID 1134.

[0069] (Figure 2: Step S106: The correction reason receiving unit receives the correction reason score from the user and stores it in the reason score storage unit.) Step S106 in Figure 2 is a process performed by the correction reason receiving unit 124 and the reason score storage unit 114. The correction reason receiving unit 124 receives the intention to correct the correction plan 1221 from the user who entered the correction plan 1221 in step S105. Specifically, the correction reason receiving unit 124 of the display unit 120 in Figure 9 receives the correction reason score 1241 from the user. Furthermore, the received correction reason score 1241 is stored in the reason score storage unit 114.

[0070] The correction reason score 1241 holds, for example, a plan ID 1141, an item 1142, a reason score 1143, and a reason score type 1144, as shown in Figure 7. For the plan ID 1141, enter the plan ID 1121 of the correction plan 1221 that the correction reason score 1241 evaluates. For the reason score type 1144, enter "Correction Reason Score".

[0071] The following describes how to receive and store item 1142 and reason score 1143.

[0072] The method for receiving item 1142 and reason score 1143 is as shown in the correction reason reception unit 124 (Figure 9), where the importance of correction reason score 1241 is entered as a positive or negative integer centered around 0, for the purpose of accepting the degree of correction. For example, if you want to increase the importance of the reason "considering proficiency" the most, you would enter "+3" for the importance of considering proficiency, and if you want to decrease the importance of the reason "leveling overtime hours", you would enter a negative number such as "-2".

[0073] In Figure 9, input is accepted via numerical radio buttons, but this can be replaced with input using sliders or natural language input using generation AI, etc. It is also possible to accept reasons for new modifications from the user in a free-text field.

[0074] The method for storing item 1142 and reason score 1143 involves taking the sum of the importance of the correction reason score 1241 received by the correction reason reception unit 124 (Figure 9) and the importance of the estimated reason score, and storing this sum in item 1142 and reason score 1143 in the reason score storage unit 114. For example, consider a case where the importance of the reason "skill level consideration" in the estimated reason score is "0", and the importance of "skill level consideration" in the correction reason score 1241 received from the user is "+3". In that case, the reason score 1143 in the reason score information 1140 stored in the reason score storage unit 114 will be stored as 0+3=3, so "3".

[0075] In the example above, the user inputs an increase or decrease in the importance of the estimated reason score, and the result of the addition or subtraction based on that input is stored as the modified reason score. However, this input method is just one example, and the user may also directly input the changed importance (i.e., the modified reason score itself to be stored in the reason score storage unit 114).

[0076] (Figure 2: Step S107: Retraining using recommendation plan, revised plan, estimated reason score, and revised reason score as input) Step S107 in Figure 2 is a process performed by the learning unit 132 and the correction difference calculation unit 133. The correction difference calculation unit 133 calculates the difference between the correction plan 1221 received from the user in step S105 and the recommendation plan 1212 generated in step S102, and the difference between the correction reason score 1241 received from the user in step S106 and the estimated reason score 1231 generated in step S102. The specific correction difference calculation flow will be described later.

[0077] The learning unit 132 generates a learning model from the difference between the correction plan 1221 and the recommendation plan 1212 obtained from the correction difference calculation unit 133, and from the difference between the correction reason score 1241 and the estimated reason score 1231. The specific learning flow will be explained next (see Figure 11). The baseline plan 1211, supply chain master information 11110, and personnel information 11120 may also be used for retraining.

[0078] Figure 11 is a flowchart illustrating the operation of the learning unit 132 of the automated plan generation system 100 in this embodiment. The steps in Figure 11 will be described below.

[0079] (Figure 11: Step S1321: Recommendation plan, revised plan, estimated reason score, and revised reason score are obtained from the plan storage unit and reason score storage unit.) Step S1321 in Figure 11 is a process performed by the plan storage unit 112 and the reason score storage unit 114. The plan storage unit 112 retrieves the recommendation plan 1212 and the modification plan 1221. Specifically, it extracts records from the plan information 1120 in Figure 5 where the plan type 1128 is a recommendation plan and records where it is a modification plan. The reason score storage unit 114 retrieves the estimated reason score 1231 and the modification reason score 1241. Specifically, it extracts records from the reason score information 1140 in Figure 7 where the reason score type 1144 is an estimated reason score and records where it is a modification reason score 1241.

[0080] (Figure 11: Step S1322: The difference between the recommendation plan and the revised plan, and the difference between the estimated reason score and the revised reason score are obtained from the revised difference calculation unit.) Step S1322 in Figure 11 is a process performed by the correction difference calculation unit 133. The correction difference calculation unit 133 obtains the difference between the recommendation plan 1212 and the correction plan 1221, and the difference between the estimated reason score 1231 and the correction reason score 1241, according to the flow shown in Figure 12, and passes them to the learning unit 132. The specific correction difference calculation flow will be described later in Figure 12.

[0081] (Figure 11: Step S1323: Learn from the difference between the recommendation plan and the revised plan, and the difference between the estimated reason score and the revised reason score, and generate a learned model.) Step S1323 in Figure 11 is a process performed by the learning unit 132. The learning unit 132 can be omitted if, as in step S1312 described in this embodiment, it calculates an estimated reason score based on the reason score information 1140 received in the correction reason score 1241 and calculates the importance to be displayed in the recommendation plan 1212 and estimated reason score 1231.

[0082] On the other hand, as a means other than the importance calculation method described in this embodiment, methods such as machine learning may be applied. In that case, the reason item relationship information 1150 corresponding to each item of the baseline plan 1211 and the revised reason score 1241 presented to the user is taken as input, and the estimated reason score and importance displayed in the recommendation plan 1212 and estimated reason score 1231 are taken as output, and the output, along with the revised plan 1221 modified by the user and the revised reason score 1241 entered by the user, are used as ground truth labels to generate a learning model.

[0083] Figure 8 is an explanatory diagram of the reason item relationship information 1150 held by the automatic plan generation system 100 of this embodiment.

[0084] As shown in Figure 8, the reason item relationship storage unit 115 of the automatic plan generation system 100 stores reason item relationship information 1150, which consists of item 1151 and input information 1152. Item 1151 corresponds to the reason item included in the correction reason score 1241. Input information 1152 indicates the information held by the automatic plan generation system 100 that corresponds to each reason item included in the correction reason score 1241.

[0085] For example, item 1151 stores information such as overtime leveling, skill level consideration, and reduction of setup changes. Among these, input information 1152, which corresponds to overtime leveling, stores the current month's overtime hours 11127 from personnel information 11120. This indicates that overtime leveling is related to the current month's overtime hours 11127 from personnel information 11120.

[0086] When using machine learning as described above, in step S1312, the trained model is used to take as input the information indicated by the input information 1152 of the reason item relationship information 1150 corresponding to each item of the baseline plan 1211 and the modified reason score 1241, and outputs the estimated reason score and importance to be displayed in the recommendation plan 1212 and estimated reason score 1231, respectively, and displays them in the recommendation plan 1212 and estimated reason score 1231.

[0087] Figure 12 is a flowchart illustrating the operation of the correction difference calculation unit 133 of the automatic plan generation system 100 in this embodiment. The following describes each step in Figure 12.

[0088] (Figure 12: Step S1331: In response to a call from the learning unit, the recommendation plan, revised plan, estimated reason score, and revised reason score are obtained from the plan storage unit and the reason score storage unit.) Step S1331 in Figure 12 is a process performed by the plan storage unit 112 and the reason score storage unit 114. The plan storage unit 112 retrieves the recommendation plan 1212 and the revised plan 1221 for which the difference is calculated, by filtering them using the plan date 1122, scheduled date 1123, and target plan 1127 of the plan storage unit 112. The reason score storage unit 114 retrieves the estimated reason score 1231 and the revised reason score 1241 for which the difference is calculated, by filtering them using the plan ID 1141, item 1142, reason score 1143, and reason score type 1144 from the reason score storage unit 114.

[0089] (Figure 12: Step S1332: Calculate the difference between the recommendation plan and the revised plan, and the difference between the estimated reason score and the revised reason score) Step S1332 in Figure 12 is a process performed by the correction difference calculation unit 133. The correction difference calculation unit 133 calculates the difference between the recommendation plan 1212 and the correction plan 1221 obtained in step S1331, and the difference between the estimated reason score 1231 and the correction reason score 1241. Specifically, the difference is obtained by subtracting one value from the other.

[0090] In calculating the difference between the recommendation plan 1212 and the revised plan 1221, the difference in the planned amount 1125 of the records extracted in step S1331 is calculated from the plan storage unit 112. In calculating the difference between the estimated reason score 1231 and the revised reason score 1241, the difference in the reason score 1143 of the records extracted in step S1331 is calculated from the reason score storage unit 114.

[0091] (Figure 12: Step S1333: Output the difference between the recommendation plan and the revised plan, and the difference between the estimated reason score and the revised reason score to the learning unit.) Step S1333 in Figure 12 is a process performed by the correction difference calculation unit 133. The correction difference calculation unit 133 outputs to the learning unit 132 the difference between the recommendation plan 1212 and the correction plan 1221 calculated in step S1332, and the difference between the estimated reason score 1231 and the correction reason score 1241.

[0092] Here, we will explain with reference to Figure 9 an example of the above process, specifically the baseline plan 1211 obtained in step S101 of Figure 2, the recommendation plan 1212 and estimated reason score 1231 generated in step S102, the revised plan 1221 received from the user in step S105, and the revised reason score 1241 received from the user in step S106.

[0093] The display unit 121 of the display unit 120 displays the base plan 1211 and the recommended plan 1212. The base plan 1211 shown in Figure 9 is an example of a base plan obtained in step S101 of Figure 2, and shows the daily production volume plan for each item on each production line of the factory. In the example in Figure 9, the daily production volume plan for items A, B, C, and D on lines A and B is shown for days 1 to 4. An actual factory may have more production lines and produce more items, and production volumes for days 5 and beyond may also be planned, but for illustrative purposes, the above plan for 2 lines, 4 items, and 4 days is shown here.

[0094] According to the baseline plan 1211 in Figure 9, it is planned to produce 25 units of item A per day on line A. This corresponds to one record of the planning information 1120 shown in Figure 5. Specifically, the record in which the scheduled date 1123 is "1 day" (for example, February 1, 2020), item ID 1124 is "Item A" (i.e., item A), planned quantity 1125 is "25", target plan 1127 is "production plan", plan type 1128 is "baseline plan", and attribute information 1129 is "line A" corresponds to the above plan.

[0095] Here, by referring to the personnel information 11120 shown in Figure 4, the number of employees in charge of line A and the skill level of each employee can be identified. Based on this information, the overtime hours required to produce 25 units of item A per day on line A can be calculated, and these overtime hours (e.g., -0.5 hours) are included and displayed in the baseline plan 1211.

[0096] Similarly, in the example of standard plan 1211 in Figure 9, on line A, 75, 50, and 50 units of item B are produced on days 2, 3, and 4, respectively, with overtime hours of 2.0 hours, 1.0 hour, and 1.0 hour. On the other hand, on line B, 60 units of item C, 40 units of item D, 70 units of item D, and 30 units of item B are produced on days 1, 2, 3, and 4, respectively, with overtime hours of 1.4 hours, 0.0 hours, 1.0 hour, and -1.0 hour.

[0097] In contrast, the recommendation plan 1212 shown in Figure 9 is an example of a recommendation plan generated in step S102 based on the base plan 1211 and input information (including, for example, the personnel information 11120 shown in Figure 3). Here, we will explain the parts of the recommendation plan 1212 that have been changed from the base plan 1211.

[0098] First, the production schedule for item B on line A has been changed so that 25 out of the 75 items produced on line A in two days are now produced on line A in one day. As a result, the overtime hours on line A are leveled out to 1.0 hour on every day from the 1st to the 4th. However, since line A now produces two items, item A and item B, per day, the number of setup changes increases.

[0099] Similarly, 10 of the 60 units of item C produced on line B in one day have been changed to be produced on line B on the second day. Furthermore, 20 of the 70 units of item D produced on line B in three days have been changed to be produced on line B on the fourth day. As a result, the overtime hours on line B are leveled out to 1.0 hour on each of the four days from the first to the fourth day. However, line B produces two items, item C and item D, on the second day, and two items, item B and item D, on the fourth day, which increases the number of setup changes.

[0100] The estimated reason display unit 123 indicates that the reason for the above changes was that the equalization of overtime hours was prioritized, while the reduction of employee skill levels and the number of setup changes was not prioritized.

[0101] The correction plan reception unit 122 displays the correction plan 1221. The correction plan 1221 shown in Figure 9 is an example of a correction plan received from the user in step S105. Here, we will explain the parts of the correction plan 1221 that have been modified from the recommendation plan 1212.

[0102] First, all 25 units of item A produced on line A in one day are now produced on line B in one day. Furthermore, 25 of the 30 units of item B produced on line B in four days are added to the daily production on line A, and 5 units are added to the four-day production on line A. 25 of the 50 units of item C produced on line B in one day are added to the two-day production on line B, and 25 of the 40 units of item D produced on line B in two days are added to the four-day production on line B.

[0103] This modification plan 1221 is an example where, because item A tends to have a high defect rate, and many highly skilled employees are assigned to line B, the user prioritized employee skill and modified the plan to move production of item A to line B and, conversely, move production of other items to line A. As a result, the importance of leveling out overtime hours decreased relatively, and consequently, the overtime hours for line A over four days increased to 1.2 hours, while the overtime hours for line B over four days decreased to 0.5 hours. Such modification reasons are entered into the modification reason reception unit 124 as described above.

[0104] If the learning unit 132 were to simply retrain the revised plan 1221 as the correct data for the recommendation plan 1212 without considering the estimated reasons and revision reasons, it would learn that item A should be produced on line B, and thereafter, recommendation plans for producing item A on line B would be more likely to be generated. In contrast, in the present invention, by retraining the revised plan 1221 and the revision reason score 1241 as the correct data for the recommendation plan 1212 and the estimated reason score 1231, respectively, it is expected that recommendation plans for producing item A on a production line with many skilled workers will be more likely to be generated. This makes it easier to generate plans that take into account the intentions of the planner, and reduces the man-hours required for planning work.

[0105] Although the above embodiments mainly described production planning, the present invention is not limited to production planning and can be applied to any business plan that constitutes a supply chain, such as procurement planning, inventory planning, transportation planning, and sales planning. Furthermore, it can be applied to any business plan, such as railway operation planning, regardless of its relationship to the supply chain.

[0106] The employees registered in personnel information 11120 and the production lines, etc., that each employee is responsible for are examples of the resources, such as personnel and equipment, that are allocated to business planning. For example, in a transportation plan, the transport vehicles and their drivers would be considered resources, and in a railway operation plan, the railway vehicles and their drivers would be considered resources.

[0107] In standard plans and similar plans, resource allocations and output quantities are planned for each period (for example, each day). Output quantities may be the production volume for each item in the case of a production plan, the transport volume in the case of a transport plan, or the number of passengers transported in the case of a railway operation plan.

[0108] The system of the embodiment of the present invention may be configured as follows, for example.

[0109] (1) A plan generation method executed by a plan generation system (e.g., an automatic plan generation system 100), wherein the plan generation system comprises a calculation unit (e.g., a calculation unit 130) and a storage unit (e.g., a storage unit 110), the storage unit containing a base plan (e.g., a base plan 1211 included in the plan information 1120) which is a plan for a business, resource information (e.g., personnel information 11120) relating to resources allocated to the business in the base plan, a recommendation plan (e.g., a recommendation plan 1212) which includes recommended changes to the base plan, and an estimated reason (e.g., an estimated reason score 123) which is an estimation of the reasons for the changes in the recommendation plan. The plan generation method includes a learning model that outputs (for example, a learning model used by the generation unit 131) and a first step (for example, step S102) in which the calculation unit generates the recommendation plan based on the reference plan, the resource information, and the learning model, and a second step (for example, steps S105 to S107) in which, when a modified plan (for example, modified plan 1221) that modifies the recommendation plan and the reason for the modification (for example, modification reason score 1241) is input, the calculation unit retrains the learning model using the modified plan and the reason for the modification as the correct data for the recommendation plan and the reason for the estimation, respectively.

[0110] This allows for the generation of plans that reflect the intentions of the planners, thereby achieving the desired plan output and reducing the workload involved in planning.

[0111] (2) The plan generation method described in (1) above, wherein the estimated reason includes one or more reason items and a score indicating the importance of each item, and the modified reason includes the amount of modification to the importance of each item included in the estimated reason, or the modified importance.

[0112] This makes it easier to understand and input the intent of the plan.

[0113] (3) The plan generation method described in (2) above, wherein the resources include personnel (e.g., employees of a factory, etc.) and equipment (e.g., a production line), the base plan includes the allocation of the resources for each period in a business constituting the supply chain (e.g., a manufacturing business) and the output of the business (e.g., the planned production quantity for each item), the resource information includes information indicating the skill level of the personnel (e.g., skill level classification 11125), information regarding the working hours of the personnel (e.g., overtime hours for the current month 11127), and information indicating the equipment that the personnel is in charge of (e.g., assigned equipment 11126), and the estimation reason and the modification reason include at least one of the following items: a reason relating to the skill level of the personnel (e.g., skill level consideration), a reason relating to the working hours of the personnel (e.g., overtime hour leveling), and a reason relating to the utilization efficiency of the equipment (e.g., reduction of changeovers).

[0114] This makes it easier to understand and input the intent of plans when generating plans for businesses that make up the supply chain.

[0115] (4) The plan generation method described in (3) above, wherein the information indicating the skill level of the personnel includes at least one of the following: the personnel's length of service (e.g., length of service 11122), qualifications held (e.g., qualification A held 11123), and evaluation from others (e.g., evaluation 11124), and information obtained from a personnel management system.

[0116] This ensures that the appropriate level of proficiency is achieved.

[0117] (5) The plan generation method described in (1) above, wherein the estimated reason and the modification reason include reasons described in natural language (e.g., free description).

[0118] This increases the degree of freedom in inputting reasons.

[0119] It should be noted that the present invention is not limited to the embodiments described above, and various modifications are included. For example, the embodiments described above are explained in detail for a better understanding of the present invention, and are not necessarily limited to those having all the configurations described. Furthermore, it is possible to replace parts of the configuration of one embodiment with the configuration of another embodiment, and it is possible to add configurations from other embodiments to the configuration of one embodiment. In addition, it is possible to add, delete, or replace parts of the configuration of each embodiment with other configurations.

[0120] Furthermore, each of the above configurations, functions, processing units, processing means, etc., may be implemented in hardware, in whole or in part, for example, by designing them as integrated circuits. Alternatively, each of the above configurations, functions, etc., may be implemented in software by having the processor interpret and execute programs that realize each function. Information such as programs, tables, files, etc., that realize each function can be stored in storage devices such as memory, hard disk drives, SSDs (Solid State Drives), or computer-readable non-temporary data storage media such as IC cards, SD cards, DVDs, etc.

[0121] Furthermore, the control lines and information lines shown are those deemed necessary for explanation purposes, and do not necessarily represent all control lines and information lines in the actual product. In practice, it can be assumed that almost all components are interconnected. [Explanation of symbols]

[0122] 100 Automatic Plan Generation System 110 Storage section 111 Input Information Storage Unit 112 Planning Memory Unit 113 Correction information storage unit 114 Reason Score Memory Unit 115 Reason Item Relationship Memory Unit 120 Display section 121 Plan Display Section 122 Revision Plan Reception Department 123 Estimated reason display section 124 Revision Reason Reception Department 130 Arithmetic section 131 Generation part 132 Learning Department 133 Modified difference calculation section 200 Communication Networks 300 locations

Claims

1. A plan generation method performed by a plan generation system, The aforementioned plan generation system comprises a calculation unit and a storage unit, The memory unit holds a baseline plan, which is a business plan; resource information relating to the resources allocated to the business in the baseline plan; a recommendation plan including recommended changes to the baseline plan; and a learning model that outputs an estimation reason, which is an estimation of the reasons for the changes in the recommendation plan. The aforementioned plan generation method is, The calculation unit performs a first step of generating the recommendation plan based on the baseline plan, the resource information, and the learning model. A plan generation method characterized by including a second step in which, upon inputting a revised plan obtained by modifying the recommendation plan and the reason for the revision, the calculation unit retrains the learning model using the revised plan and the reason for the revision as the correct data for the recommendation plan and the reason for the estimation, respectively.

2. A plan generation method according to claim 1, The aforementioned reasoning includes one or more reason items and a score indicating the importance of each item. A plan generation method characterized in that the aforementioned reasons for modification include the amount of modification to the importance of each item included in the aforementioned reasons for modification, or the importance after modification.

3. A plan generation method according to claim 2, The aforementioned resources include personnel and equipment, The aforementioned baseline plan includes the allocation of resources and the output of the businesses that constitute the supply chain at different time periods. The resource information includes information indicating the skill level of the personnel, information regarding the working hours of the personnel, and information indicating the equipment that the personnel are responsible for. A plan generation method characterized in that the aforementioned estimation reasons and revision reasons include at least one of the following items: reasons relating to the skill level of the personnel, reasons relating to the working hours of the personnel, and reasons relating to the utilization efficiency of the equipment.

4. A plan generation method according to claim 3, A plan generation method characterized in that the information indicating the skill level of the personnel includes at least one of the following: information based on the personnel's length of service, qualifications held, and evaluations from others, and information obtained from a personnel management system.

5. A plan generation method according to claim 1, A plan generation method characterized in that the aforementioned estimation reasons and revision reasons include reasons described in natural language.

6. A plan generation system, It comprises an arithmetic unit and a memory unit, The memory unit holds a baseline plan, which is a business plan; resource information relating to the resources allocated to the business in the baseline plan; a recommendation plan including recommended changes to the baseline plan; and a learning model that outputs an estimation reason, which is an estimation of the reasons for the changes in the recommendation plan. The aforementioned arithmetic unit, Based on the aforementioned baseline plan, the resource information, and the learning model, the recommendation plan is generated. A plan generation system characterized in that, when a revised plan obtained by modifying the recommendation plan and the reason for the revision are input, the system retrains the learning model using the revised plan and the reason for the revision as the correct data for the recommendation plan and the reason for the estimation.

Images