JP6696655B2 - Plan making program, plan making apparatus, and plan making method - Google Patents

Description

  The present invention relates to a plan creation program and the like for creating a business plan.

  Conventionally, a system that supports the creation of a business plan for a company has been proposed (Patent Document 1).

JP, 2004-334620, A

  However, to create a business plan using the conventional system, the income statement (P / L), balance sheet (B / S), cash flow statement (C / F), etc. for the past few years are used. It is complicated because it is necessary to input actual data. In addition, it is necessary to confirm many index values such as sales and cost of sales that constitute the management plan before the management plan is finalized, but it is difficult to understand in the list display. The present invention has been made in view of such circumstances. Its purpose is to provide a plan creation program or the like that collects the actual data necessary for creating the business plan and outputs a graph representing the created business plan.

The plan creation program according to the present invention acquires the value of each of the plurality of description items for a plurality of main items including sales and cost of sales related to management, and when the values of the plurality of description items are input, To the learned model that outputs the predicted values of the plurality of main items, input the values of each of the acquired plurality of description items, and obtain the predicted values of the plurality of main items that the learned model outputs , Output the question to select and answer the probability of variation about the item, acquire the selection information of the probability of variation for the question, correlate with the acquired selection information, and convert the coefficient obtained by converting the probability of variation. Acquiring a calculation logic including, correcting the predicted value of the main item based on the acquired calculation logic, and based on the corrected predicted value of the main item, at least a sales or gross profit configuring the management plan It is characterized in that the computer is caused to perform a process of calculating an index value including the two and outputting the created index value in a graph displayable manner.

  In the present invention, the values of the explanation items for the main items necessary for creating the business plan are collected, and the values of the collected explanation items are input to the learned model to create the predicted values of the main items. It is possible to output the management plan that was created as a graph.

It is explanatory drawing which shows the structural example of a plan management system. It is a block diagram which shows the hardware structural example of a management server. It is a block diagram which shows the hardware structural example of a user terminal. It is a block diagram showing an example of hardware constitutions of an administrator terminal. It is explanatory drawing which shows the example of user DB. It is explanatory drawing which shows the example of accounting DB. It is explanatory drawing which shows the example of accounting DB. It is explanatory drawing which shows the example of reservation DB. It is explanatory drawing which shows the example of POS-DB. It is explanatory drawing which shows the example of attendance DB. It is explanatory drawing which shows the example of order receipt / order DB. It is explanatory drawing which shows the example of logic DB. It is explanatory drawing which shows the example of a prediction DB. It is explanatory drawing which shows the example of report DB. It is explanatory drawing which shows the example of alert DB. It is a flow chart which shows the example of a procedure of plan creation processing. It is a flow chart which shows the example of a procedure of automatic creation processing. It is an explanatory view showing an example of automatic creation. It is a flow chart which shows the example of a procedure of question creation processing. It is a flow chart which shows the example of a procedure of question creation processing. It is explanatory drawing which shows the example of question preparation. It is a flow chart which shows the example of a procedure of direct edit processing. It is a flow chart which shows the example of a procedure of graph edit processing. It is a flow chart which shows the example of a procedure of simulation edit processing. It is an explanatory view showing an example of a creation method selection screen. It is an explanatory view showing an example of an automatic creation setting screen. It is explanatory drawing which shows the example of a question answer screen. It is explanatory drawing which shows the example of a plan display screen. It is explanatory drawing which shows the example of an edit method selection screen. It is explanatory drawing which shows the example of a direct edit screen. It is explanatory drawing which shows the example of a graph edit screen. It is explanatory drawing which shows the example of a simulation edit screen. It is a flow chart which shows the example of a procedure of planned operation processing. It is a flow chart which shows the example of a procedure of a preliminary and actual management processing. It is a flow chart which shows the example of a procedure of report setting processing. It is a flow chart which shows the example of a procedure of alert setting processing. It is a flow chart which shows the example of a procedure of batch processing. It is a flow chart which shows the example of a procedure of preliminary and actual creation processing. It is a flow chart which shows the example of a procedure of report creation processing. It is explanatory drawing which shows the example of an automatic report creation logic. It is a flow chart which shows the example of a procedure of alert creation processing. It is a flow chart which shows the example of a procedure of prediction / re-creation processing. It is explanatory drawing which shows the example of a prediction logic. It is explanatory drawing which shows the example of an operation menu screen. It is explanatory drawing which shows the example of a forecast actual management menu screen. It is an explanatory view showing an example of a preliminary and actual management setting screen. It is explanatory drawing which shows the example of a prediction actual condition display screen. It is explanatory drawing which shows the example of a report selection screen. It is explanatory drawing which shows the example of a fixed form report setting screen. It is explanatory drawing which shows the example of a report setting screen. It is explanatory drawing which shows the example of an automatic report. It is explanatory drawing which shows the example of an alert setting screen. It is explanatory drawing which shows the example of a prediction / re-creation menu screen. It is explanatory drawing which shows the example of the reproduction method selection screen. It is explanatory drawing which shows the example of advice DB. It is explanatory drawing which shows the example of advertisement DB. It is a flowchart which shows the other example of a procedure of a prediction actual preparation process and an alert preparation process. It is explanatory drawing which shows the example of a cooperation setting screen. It is explanatory drawing which shows the multi CSV upload function. It is explanatory drawing which shows the example of collection item DB. It is explanatory drawing which shows the example of sales DB. It is explanatory drawing regarding a sales forecast model. It is a flow chart which shows the example of a procedure of learning / prediction processing. It is a flow chart which shows the example of a procedure of generation processing of a prediction model. It is a flow chart which shows the example of a procedure of prediction processing. It is explanatory drawing which shows the example of sales DB. It is explanatory drawing regarding the production | generation process of a prediction model. It is a flow chart which shows the example of a procedure of prediction processing. It is explanatory drawing which shows the example of preparation of sales forecast. It is explanatory drawing which shows the example of preparation of sales cost forecast. It is explanatory drawing which shows the example of labor cost estimation. It is explanatory drawing which shows the example of P / L based on a predicted value. It is a flow chart which shows the example of a procedure of planned value calculation processing. It is explanatory drawing which shows the example of a prediction setting screen. It is a flowchart which shows the example of a procedure of a prediction setting process. It is explanatory drawing which shows the example of external environment information DB. It is a flowchart which shows the other example of a procedure of a plan value calculation process. It is a flow chart which shows the example of a procedure of external environment information creation processing. It is explanatory drawing which shows the example which produces | generates external environment information from external information. It is explanatory drawing which shows the correction example by external environment information.

  Hereinafter, embodiments will be described with reference to the drawings.

Embodiment 1
FIG. 1 is an explanatory diagram showing a configuration example of a plan management system. The plan management system 100 includes a management server (plan creation device) 1, a user terminal 2 and an administrator terminal 3. The plan management system 100 is connected to the other system 4 which is various data sources. The management server 1, the user terminal 2, the administrator terminal 3 and the other system 4 are communicably connected to each other via a network N. The other system 4 includes a reservation system, a POS (Point of sale system) system, an attendance system, an accounting system, a salary system, and an ordering system. Reservation system includes medical appointments for medical institutions, seat reservations for restaurants, facility reservations for rental conference rooms, tour reservations for travel, ticket reservations for transportation, lesson reservations for language learning, treatment reservations for beauty salons and nail salons, etc. Including a system for making reservations. The POS system includes not only a general system that totals the sales results of goods sales in units of individual products but also a system that totals the sales results of services in the service industry in units of service menus. The attendance system is a system that manages attendance of employees including regular employees, contract employees, and part-timers. The accounting system is a management accounting system, a financial accounting system, or the like. The salary system is a system for calculating salaries to be paid to employees and storing a history of salary payments. The order receiving / ordering system is a system for receiving and ordering goods and services. In FIG. 1, two user terminals 2 and two administrator terminals 3 are shown, but each may be one, or three or more. One user terminal 2 and one administrator terminal 3 may be combined. The other system 4 may further include a purchase inventory management system, an employee goal / evaluation management system, an employee motivation management system, a word-of-mouth site management system, a coupon site management system, a customer questionnaire management system, etc. Good.

  FIG. 2 is a block diagram showing a hardware configuration example of the management server. The management server 1 collects various data relating to the management of each user from the other system 4 from the other system 4 and performs a process of creating a management plan based on the collected data. The management server 1 is composed of a server computer or the like. The management server 1 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a large-capacity storage unit 14, and a communication unit 15. The components are connected by a bus B.

  The CPU 11 controls each part of the hardware according to the control program (plan creation program) 1P stored in the ROM 12. The RAM 13 is, for example, SRAM (Static RAM), DRAM (Dynamic RAM), or flash memory. The RAM 13 temporarily stores data generated when the CPU 11 executes the program.

  The large-capacity storage unit 14 is, for example, a hard disk or an SSD (Solid State Drive). The large-capacity storage unit 14 stores various databases (DB: Data Base). The large-capacity storage unit 14 stores a user DB 141, an accounting DB 142, a reservation DB 143, a POS-DB 144, an attendance DB 145, an ordering and placing DB 146, a logic DB 147, a forecast DB 148, a report DB 149, and an alert DB 14A. Further, the control program 1P may be stored in the mass storage unit 14. All or a part of the databases from the user DB 141 to the alert DB 14A may be stored in other than the management server 1. For example, it may be stored in a database server or cloud storage.

  The communication unit 15 communicates with the user terminal 2, the administrator terminal 3 and the other system 4 via the network N. Further, the CPU 11 may use the communication unit 15 and the management server 1 may be composed of a plurality of physical server computers. Further, the function of the management server 1 may be provided by a cloud service.

  The reading unit 16 reads the portable storage medium 1a including a CD (Compact Disc) -ROM and a DVD (Digital Versatile Disc) -ROM. The CPU 11 may read the control program 1P from the portable storage medium 1a via the reading unit 16 and store it in the large-capacity storage unit 14. Further, the CPU 11 may download the control program 1P from another computer via the network N or the like and store it in the mass storage unit 14. Furthermore, the CPU 11 may read the control program 1P from the semiconductor memory 1b.

  FIG. 3 is a block diagram showing a hardware configuration example of a user terminal. The user terminal 2 is a terminal used by a user of the plan management system 100. The user terminal 2 is composed of a notebook computer, a panel computer, a tablet computer or the like. The user terminal 2 includes a CPU 21, a ROM 22, a RAM 23, a communication unit 24, an input unit 25, a display unit 26, and a large capacity storage unit 27. The components are connected by a bus B.

  The CPU 21 controls each part of the hardware according to the control program (plan editing program) 2P stored in the ROM 22. The RAM 23 is, for example, SRAM, DRAM or flash memory. The RAM 23 temporarily stores the data generated when the CPU 21 executes the program.

  The communication unit 24 communicates with the management server 1 and the other system 4 via the network N. Further, the CPU 21 may use the communication unit 24 to download the control program 2P from another computer via the network N or the like and store the control program 2P in the large capacity storage unit 27.

  The display unit 26 includes a liquid crystal display panel and the like. The display unit 26 displays the plan output by the management server 1. The input unit 25 is a keyboard or a mouse. The input unit 25 may be a touch panel display integrated with the display unit 26. The user terminal 2 may display on an external display device.

  The large-capacity storage unit 27 is, for example, a hard disk or SSD. The large capacity storage unit 27 stores various data.

  FIG. 4 is a block diagram showing a hardware configuration example of the administrator terminal. The administrator terminal 3 is a terminal used by the administrator of the plan management system 100. The administrator terminal 3 is composed of a notebook computer, a panel computer, a tablet computer or the like. The administrator terminal 3 includes a CPU 31, a ROM 32, a RAM 33, a communication unit 34, an input unit 35, a display unit 36, and a large capacity storage unit 37. The components are connected by a bus B. Since each configuration of the administrator terminal 3 is the same as each configuration of the user terminal 2 having the same name, description thereof will be omitted.

  Next, the database used by the plan management system 100 will be described. FIG. 5 is an explanatory diagram showing an example of the user DB. The user DB 141 stores information on users who use the plan management system 100. The user DB 141 includes a user ID column, a form column, a name / name column, a telephone number column, a mail address column, a password column, a chat column, an account column, and a cooperation setting column. The user ID column stores a user ID that uniquely identifies the user. The user ID is issued at the time of user registration. The name / name column stores the user's name or name. When the user is a company, the company name is stored in the name / name column. If the user is a sole proprietor, the personal name is stored in the name / name column. The telephone number column stores the telephone number of the user. The email address column stores the email address of the user. The password string stores the login password of the user. The encrypted password is stored in the password column. In FIG. 5, it is indicated by “*” for convenience. The chat column stores the identification information of the chat service that the user wants to receive the alert, for example, the name of the chat service. The account column stores the user's account in the chat service. The cooperation setting column stores the cooperation setting with the other system 4. The cooperation setting column includes an accounting column, a POS column, an attendance column, and a reservation column. The accounting column, the POS column, the attendance column, and the reservation column each store identification information of the accounting system, the POS system, the attendance system, and the appointment system, for example, a name.

  6 and 7 are explanatory diagrams showing an example of the accounting DB. The accounting DB 142 mainly stores accounting data of users collected from the accounting system. As shown in FIG. 6, the accounting DB 142 stores annual data / past 1421 and annual data / current period 1422. The annual data / past 1421 stores past data on a yearly basis. For example, annual data / past 1421 includes P / L plan 14211, P / L result 14212, B / S plan 14213, B / S result 14214, C / F plan 14215, C / F result 14216, sales plan 14217, And sales record 14218. The P / L plan 14211 includes several years of P / L plan data for past years. The P / L performance 14212 includes P / L performance data for past years for several years. The B / S plan 14213 includes B / S plan data for past years for several years. The B / S performance 14214 includes performance data for past years for several years. The C / F plan 14215 includes C / F plan data for past years for several years. The C / F performance 14216 includes C / F performance data for past years for several years. Sales plan 14217 includes sales plan data for past years for several years. The sales record 14218 includes sales record data for past years for several years. The annual data / current period 1422 includes, for example, a P / L plan 14221, a B / S plan 14222, a C / F plan 14223, and a sales plan 14224. The P / L plan 14221 is P / L plan data for the current period. The B / S plan 14222 is B / S plan data for the current period. The C / F plan 14223 is C / F plan data for the current period. Sales plan 14224 is sales plan data for the current period. The sales of the sales plan include sales of each product or service, sales of each product group or service group in which a plurality of products or services are collected, sales of each customer, and the like. In addition, one year may be the year of the calendar beginning January and ending December, or the year beginning April and ending the following March. Further, the start month and the end month may be set according to the fiscal year of the user. The past data may be P / L, B / S, C / F, and past plan and sales data regarding sales may be stored not only on a yearly basis but on a monthly basis.

  Further, as shown in FIG. 7, the accounting DB 142 stores monthly data / plan 1423 and monthly data / actual results 1424. The monthly data / plan 1423 and the monthly data / actual result 1424 are data for the current period. The monthly data / plan 1423 includes, for example, a P / L plan 14231, a B / S plan 14232, a C / F plan 14233, and a sales plan 14234. The P / L plan 14231 includes monthly P / L plan data for the current period. The B / S plan 14232 includes monthly B / S plan data for the current period. The C / F plan 14233 includes monthly C / F plan data for the current period. Sales plan 14234 includes plan data for monthly sales for the current period. The monthly data / actual result 1424 includes a P / L result 14241, a B / S plan 14242, a C / F plan 14243, and a sales plan 14244. The P / L performance 14241 stores P / L performance data for the current period on a monthly basis. The B / S plan 14242 stores the B / S performance data for the current period on a monthly basis. The C / F plan 14243 stores the C / F performance data for the current period on a monthly basis. The sales plan 14244 stores the actual sales data for the current period on a monthly basis. For example, if the current period is 2018 and the current month is June, the monthly data / actual results 1424 include complete data for April and May 2018, and the June data are incomplete in the middle of the month. It contains various data.

  FIG. 8 is an explanatory diagram showing an example of the reservation DB. The reservation DB 143 includes initial registration data 1431 and real-time data 1432. The initial registration data 1431 is static data. The initial registration data 1431 is defined for each industry. The initial registration data 1431 of the restaurant includes, for example, a user ID, a shop ID, seat information, the number of frames, the number of reservation plans, and the like. The seat information includes the number of tables and the number of seats. A frame is a time frame that includes start and end times. The number of frames is the number of frames per day. The number of reservation plans is the number of plans including service contents and charges that the customer can select at the time of reservation. The initial registration data 1431 of the massage salon includes, for example, a user ID, a shop ID, the number of treatments, the number of frames, and the number of treatment plans. The number of treatments is the number of treatment tables, and indicates the maximum number of customers who can receive treatments at the same time. The number of frames is the same as for restaurants. The number of treatment plans is the number of plans including the treatment content, treatment time, and treatment fee. The real-time data 1432 collects the reservation status of each store as needed. The real-time data 1432 includes, for example, a user ID, a store ID, a data update date / time, the number of reservations, and a cancellation rate. The cancellation rate is the number of reservations canceled on the day of business divided by the number of reservations.

  FIG. 9 is an explanatory diagram showing an example of the POS-DB. The POS-DB 144 stores data collected from the POS system as needed. The POS-DB 144 includes sales ranking 1441, cost ranking 1442, customer unit price 1443, repeat rate 1444, number of coupons 1445, consumption material 1446, and the like. The repeat rate 1444 includes daily, weekly, and monthly repeat rates. How many times the repeat rate is calculated is determined appropriately according to the type of industry. The coupon number 1445 includes the type and number of issued coupons, the type and number of used coupons, and the like. The consumed material 1446 includes the name and quantity of the consumed material.

  FIG. 10 is an explanatory diagram showing an example of the attendance DB. The attendance DB 145 stores attendance data collected from the attendance system at any time. The attendance data includes real time data 1451 and monthly data 1452. The real-time data 1451 includes attendance / leaving data for each employee. The real-time data 1451 includes, for example, time and attendance distinction, name, time and date of attendance, cumulative working hours for each month, number of working days for each month, and number of consecutive working days for the current day. The monthly data 1452 includes monthly attendance data. The monthly data 1452 may include data for each store in addition to the data obtained by summing the real-time data 1451 for each month. The monthly data 1452 for each store includes, for example, the store name, the data monthly, the total number of employees, the number of full-time employees, the number of part-time jobs, the number of new employees, and the number of retirees.

  FIG. 11 is an explanatory diagram showing an example of the ordering / ordering DB. The ordering / ordering DB 146 stores ordering / ordering data collected from the ordering / ordering system at any time. The order data 1461 includes data monthly, customer code, order product name, order unit price, and order quantity. The order data 1461 of the customer who has received an order for a plurality of products may repeatedly include the name of the ordered product, the unit price of the ordered product, and the quantity of the ordered product in one record. The order data 1462 includes data monthly, supplier code, order product name, order unit price, and order quantity. The order data 1462 of the supplier who has ordered a plurality of products may repeatedly include the ordered product name, the ordered unit price, and the ordered quantity in one record. The data stored in the reservation DB 143, the POS-DB 144, the attendance DB 145, and the ordering / ordering DB 146 may be not only the data generated in the current period but also the data in the past period. Further, the current period data and the past period data collected for each day, weekly, monthly, and yearly may be stored. In addition, plan data for the current period and past periods may be stored.

  FIG. 12 is an explanatory diagram showing an example of the logic DB. The logic DB 147 stores the logic for planning and forecasting. FIG. 12A shows an example of automatic creation logic 1471. The automatic creation logic 1471 shown in FIG. 12A is a logic for making a P / L plan for the current term from past P / L data. FIG. 12B shows an example of the query logic 1472. For example, when the user replies “slightly increase” with respect to the expected change in selling expenses and general administrative expenses, the planned value for this term is calculated to increase by 5% from the previous term (previous term × 1.05). FIG. 12C shows an example of the prediction logic 1473. The prediction logic 1473 illustrated in FIG. 12C is a logic that predicts annual sales from January to December and monthly sales from April to December using the sales results from January to March. ..

  FIG. 13 is an explanatory diagram showing an example of the preliminary and actual DB. The actual DB 148 stores the numerical values of the actual management target. The actual DB 148 includes a user ID column, a store ID column, an item column, a January column, a February column, a March column, ..., And a December column. The user ID column stores the user ID. The store ID column stores store IDs. The item string stores the management target. For example, sales, cost of sales, SG & A expenses, etc. The January column, the February column, the March column, ..., And the December column store the plan value, the actual value, and the difference value between the two values for each month.

  FIG. 14 is an explanatory diagram showing an example of the report DB. The report DB 149 stores the setting of the forecast management report. The report DB 149 includes a user ID column, a unit column, a subject column, an annual plan column, a cumulative value column, a format column and an image column. The user ID column stores the user ID. The unit column stores units for displaying numerical values. The unit is a store, region, business unit, company-wide, etc. The item string stores the subject of the actual management. The yearly planning column stores whether or not to display the yearly planned value in the report. The cumulative value column stores whether or not to display the cumulative value up to this month in the report. For example, when the cumulative sales value is displayed in the April report, the total sales value from January to April is displayed. The format column stores the format name when the report is created. The format is stored in advance. The image column stores the latest report as an image or PDF.

  FIG. 15 is an explanatory diagram showing an example of the alert DB. The alert DB 14A stores alert conditions set by the user. The alert DB 14A includes a user ID column, a number column, an evaluation index column, a reference value column, a unit column, an alert method column, and an address column. The user ID column sets alert conditions and stores the user ID of the user who receives the alert. The number column stores the sequential number of the alert condition. The sequence number is unique to each user. The evaluation index column stores indexes (subjects) to be evaluated. The evaluation index is, for example, sales, gross profit, or customer unit price. The reference value column stores reference values for determining whether or not to perform an alert. The unit column stores the unit of the reference value. The unit is, for example,%, one million yen, or ten million yen. The alert method column stores the alert method. As an alert method, for example, an email, an SMS (Short Message Service) address column stores an address to which an alert is sent. When the alert method is email, the address column stores the email address.

  The plan management system 100 can be utilized at the time of creating and operating a business plan. First, a process performed by the plan management system 100 when a user creates a plan using the plan management system 100 will be described.

  FIG. 16 is a flowchart showing an example of the procedure of the planning process. The CPU 11 transmits the creation method selection screen to the user terminal 2. The user selects a creation method on the creation method selection screen. The creation method selected by the user is transmitted to the management server 1. The CPU 11 of the management server 1 receives the plan creation method from the user terminal 2 (step S1). The CPU 11 determines whether the acquired creation method is automatic creation (step S2). When the CPU 11 determines that the acquired creation method is automatic creation (YES in step S2), it automatically creates (step S3). When the CPU 11 determines that the acquired creation method is not automatic creation (NO in step S2), it creates a question (step S4). The CPU 11 transmits the plan created in step S2 or S3 to the user terminal 2 (step S5). The user terminal 2 acquires and displays the plan received from the management server 1. The user refers to the displayed plan, determines whether to accept the plan as it is or edits it, and transmits the result from the user terminal 2 to the management server 1. The CPU 11 of the management server 1 receives the determination result from the user terminal 2 (step S6). The CPU 11 determines whether the determination result received from the user terminal 2 is edit (step S7). When the CPU 11 determines that the determination result is not editing (NO in step S7), the process proceeds to step S13. When the CPU 11 determines that the determination result is editing (YES in step S7), it determines whether the editing method is direct input (step S8). When the CPU 11 determines that the editing method is direct input (YES in step S8), it directly edits (step S9). When the CPU 11 determines that the editing method is not direct input (NO in step S8), it determines whether it is graph editing (step S10). When the CPU 11 determines that the editing method is graph editing (YES in step S10), it performs graph editing (step S11). When the CPU 11 determines that the editing method is not graph editing (NO in step S10), it performs simulation editing (step S12). The CPU 11 stores the created plan or the edited plan (step S13). The database that stores the plan depends on the content. For example, when the created plan is P / L, it is stored in the accounting DB 142. It should be noted that when a plan for a plurality of periods is created, the series of processes shown in FIG. 16 is repeatedly executed.

  FIG. 17 is a flowchart showing an example of the procedure of automatic creation processing. The automatic creation process is a process of estimating a change from the previous period and creating a plan for the current period based on actual data. The CPU 11 of the management server 1 transmits the setting screen to the user terminal 2 (step S21). The CPU 21 of the user terminal 2 displays the setting screen on the display unit 26 (step S22). The user terminal 2 uses the input unit 25 to set the method for acquiring the actual result data and instruct the transmission. The CPU 21 transmits the acquisition method of the result data to the management server 1 (step S23). The CPU 11 of the management server 1 determines whether the acquisition method is file upload (step S24). When the CPU 11 determines that the acquisition method is file upload (YES in step S24), it transmits the file setting screen (step S25). The CPU 21 of the user terminal 2 displays the file setting screen on the display unit 26 (step S26). The user uses the input unit 25 to set a file to be uploaded and gives an instruction for transmission. The CPU 21 sends the file to the management server 1 (step S27). The CPU 11 of the management server 1 acquires the performance data (step S31). When the CPU 11 determines that the acquisition method is not file upload (NO in step S24), the CPU 11 transmits the cooperation setting screen to the user terminal 2 (step S28). The CPU 21 of the user terminal 2 displays the cooperation setting screen on the display unit 26 (step S29). The user uses the input unit 25 to set the data cooperation with the other system 4. The CPU 21 transmits the cooperation setting to the management server 1 (step S30). The CPU 11 of the management server 1 acquires the actual result data from the other system 4 according to the cooperation setting (step S31). The CPU 11 transmits the acquisition completion of the result data to the user terminal 2 (step S32). The user sets the automatic creation and gives an instruction to send. The CPU 21 transmits the setting to the management server 1 (step S33). The CPU 11 of the management server 1 acquires the settings (step S34). The CPU 11 creates a plan according to the acquired settings and the automatic creation logic stored in the logic DB 147 (step S35). The CPU 11 returns the process to the calling source. Note that step S25, step S26, step S27, and step S31 may be repeatedly executed to upload the file from the user terminal 2 to the management server 1 in multiple times.

  FIG. 18 is an explanatory diagram showing an example of automatic creation. The process described with reference to FIG. 18 corresponds to step S35 of FIG. 17, and is an example of the process performed by the CPU 11 of the management server 1. FIG. 18 shows an example of creating P / L data. According to the automatic creation logic 1471 shown in FIG. 12A, this is an example of creating a plan for this term (2018) by using the P / L performance data of the past two terms (2016, 2017). The variable cost ratio is not included in P / L, but it is necessary for planning. The table in the upper part of FIG. 18 is an example in which the increase / decrease rate is calculated from the data of the past two periods for the P / L subject and the variable cost ratio. This is the preparation process for applying the automatic creation logic. The table in the lower part of FIG. 18 shows an example of creating a plan for this term (2018) by the automatic creation logic. Sales are the sales in 2017 plus the rate of change in the past two periods. That is, it is assumed that the rate of increase / decrease from 2016 to 2017 and the rate of increase / decrease from 2017 to 2018 are the same. The cost of sales is calculated by multiplying the sales by the variable cost ratio. Here, it is assumed that the variable cost ratio in 2018 does not change from 2017. Similar to the sales amount, the rate of change may be the same as the rate of change in the past two periods. Gross profit is calculated by subtracting cost of sales from sales. It is assumed that selling, general and administrative expenses fluctuate to a value of 2017 at the same rate of change as in the past two periods. Gross profit is obtained by subtracting selling, general and administrative expenses from gross profit.

  Although the P / L plan is taken as an example, the B / S, C / F, sales plan, and the like can also be similarly planned for the current period from the past two periods of actual data. Furthermore, not only the annual plan but also the monthly plan can be planned according to the same logic.

  19 and 20 are flowcharts showing an example of the procedure of the question creating process. The question creation process is a process of estimating a change from the previous term based on the answer to the question to the user and creating a plan for the current term. In FIG. 19, steps S41 to S52 are the same as steps S21 to S42 of FIG. 18, so description thereof will be omitted. After step S52, the user makes settings for question creation and instructs transmission. The CPU 21 sends the setting to the management server 1 (step S53). The CPU 11 of the management server 1 acquires the settings (step S54). The CPU 11 acquires a question for the user from the logic DB 147 based on the acquired settings (step S55). The CPU 11 transmits a question screen including the acquired question to the user terminal 2 (step S56). The CPU 21 of the user terminal 2 displays the question screen on the display unit 26 (step S57). The user uses the input unit 25 to input an answer to the question and instructs the transmission. The CPU 21 transmits the answer to the management server 1 (step S58). The CPU 11 of the management server 1 acquires the answer (step S59). The CPU 11 determines whether there are remaining questions and remaining questions (step S60). When the CPU 11 determines that there are remaining questions (YES in step S60), the process returns to step S56. When the CPU 11 determines that there are no remaining questions (NO in step S60), it creates a plan based on the answers (step S61). The CPU 11 returns the process to the calling source.

  FIG. 21 is an explanatory diagram showing an example of creating a question. The process described with reference to FIG. 21 corresponds to step S61 of FIG. 20, and is an example of the process performed by the CPU 11 of the management server 1. FIG. 21 shows an example of creating P / L data. FIG. 12B is an example of creating a plan for this term (2018) from the data for the previous term (2017) according to the query logic 1472. Again, the variable cost ratio is not included in P / L, but it is necessary for planning. The table in the upper part of FIG. 21 is an example in which the planned numbers for the current term are calculated for some P / L subjects based on the user's answers to the questions. The user was asked the question "How much would you like to increase the selling expenses and general administrative expenses?", And the answer was "slightly increased". According to the query logic 1472, “slight increase” is replaced with 5% increase. Then, based on the figures for the previous term, we are looking for the planned figures for this term. In addition, he asked the user "How much will the target operating profit increase?" And obtained "20%" as the answer. Based on this answer, the operating profit for the current term is expected to increase by 20%, and the target value of the gross operating profit for the current term is sought. Further, regarding the variation ratio, the user is asked a question "how much will it decrease?", And an answer of "slightly decrease" is obtained as the answer. According to the query logic 1472, “slight decrease” is replaced with 2% decrease. Then, based on the figures for the previous term, we are seeking the planned figures for this term. The lower part of FIG. 21 obtains sales, cost of sales, and gross profit from the selling, general and administrative expenses, operating profit, and variable cost ratio obtained from the answers. Sales for the current fiscal year can be calculated from the selling, general and administrative expenses, operating income and variable cost ratio obtained for the current fiscal period. Cost of sales can be calculated from sales and variable cost ratio. Gross profit is calculated from sales and cost of sales.

  Although the P / L plan is taken as an example, B / S, C / F, sales plan, etc. similarly set questions and selected answers for some subjects, and set the planned values according to the answers from the user. Ask. Then, it is possible to obtain the planned values for the remaining subjects based on the obtained planned values. Further, not only the annual plan but also the monthly plan can be planned by the same question logic. When creating a plan for multiple periods, first create a plan for this period. After that, a plan will be created in order for next term, next term, and so on.

  FIG. 22 is a flowchart showing an example of the procedure of direct editing processing. The direct editing process is a process of directly editing the numerical value included in the plan to edit the plan automatically created or the question created. The CPU 11 of the management server 1 transmits the edit screen to the user terminal 2 (step S71). The CPU 21 of the user terminal 2 displays the edit screen on the display unit 26 (step S72). The user uses the input unit 25 to change the planned value displayed on the edit screen as necessary. The plan value may be changed by directly entering a numerical value or a mathematical expression. For example, when the planned value is 110, if = 110 * 0.85 is input, the planned value can be changed to the original value of 85%. The CPU 21 determines whether the planned value has been changed (step S73). When the CPU 21 determines that the planned value has been changed (YES in step S73), it redisplays the screen (step S74). At this time, other planned values linked with the changed planned value are also updated. When the CPU 21 determines that the planned value has not been changed (NO in step S73), the process proceeds to step S75. The CPU 21 determines whether to decide the plan (step S75). The CPU 21 determines whether or not the user has input an instruction to finish editing and decide a plan. When the CPU 21 determines that the plan is not decided (NO in step S75), the process returns to step S73. When determining that the plan is to be determined (YES in step S75), the CPU 21 transmits the editing result to the management server 1 (step S76). The CPU 11 of the management server 1 acquires the editing result (step S77). The CPU 11 returns the process to the calling source. The edit result transmitted from the user terminal 2 to the management server 1 may be the correspondence between the course and the plan value after the change, only for the course for which the plan value has been changed, or all of them including those not changed. The planned value of may be transmitted.

  FIG. 23 is a flowchart showing an example of the procedure of the graph editing process. The graph editing process is a process of displaying a plurality of numerical values included in the plan in a graph and editing the displayed graph to edit the created plan. The CPU 11 of the management server 1 transmits the edit screen to the user terminal 2 (step S81). The CPU 21 of the user terminal 2 displays the edit screen on the display unit 26 (step S82). The user changes the graph by operating the graph displayed on the edit screen by the input unit 25, particularly a pointing device such as a mouse or a trackpad, as necessary. The CPU 21 determines whether the graph has been changed (step S83). When the CPU 21 determines that the graph has not been changed (NO in step S83), the process proceeds to step S88. When the CPU 21 determines that the graph has been changed (YES in step S83), the CPU 21 acquires the graph to be changed and the change content (step S84). The CPU 21 converts the changed contents of the graph into numerical values (step S85). The CPU 21 changes the corresponding planned value based on the converted numerical value (step S86). The CPU 21 redisplays the screen (step S87). The CPU 21 determines whether to decide the plan (step S88). It is determined whether or not the user has finished editing and has input an instruction to determine the plan. When the CPU 21 determines that the plan is not decided (NO in step S88), the process returns to step S83. When the CPU 21 determines to determine the plan (YES in step S88), it sends the edit result to the management server 1 (step S89). The CPU 11 of the management server 1 acquires the editing result (step S90). The CPU 11 returns the process to the calling source. The editing result transmitted from the user terminal 2 to the management server 1 is the same as in the case of direct editing.

  FIG. 24 is a flowchart showing a procedure example of the simulation editing process. The simulation editing process is a process of editing the created plan on the assumption that a predetermined numerical value included in the plan has changed. The CPU 11 of the management server 1 transmits the setting screen to the user terminal 2 (step S101). The CPU 21 of the user terminal 2 displays the setting screen on the display unit 26 (step S102). The user uses the input unit 25 to set simulation conditions. For example, the sales amount is increased by 10% from the current planned value, and the variable cost ratio is decreased by 1 point from the planned value. The CPU 21 determines whether to determine the condition (step S103). It is determined whether or not the user has finished the setting and has input an instruction to determine the condition. When the CPU 21 determines that the condition is not determined (NO in step S103), step S103 is repeatedly executed. When determining that the condition is to be determined (YES in step S103), the CPU 21 transmits the setting content of the simulation condition to the management server 1 (step S104). The CPU 11 of the management server 1 acquires the setting contents (step S105). The CPU 11 recalculates the planned value according to the set simulation conditions (step S106). The CPU 11 returns the process to the calling source.

  Next, the screen displayed on the user terminal 2 in the above-described processing will be described. FIG. 25 is an explanatory diagram showing an example of a creation method selection screen. The creation method selection screen d01 is a screen for selecting a method for creating a plan. The creation method selection screen d01 includes an automatic creation button d011 and a question creation button d012. When the automatic creation button d011 is selected, the management server 1 creates a plan by automatic creation. When the question creation button d012 is selected, the management server 1 creates a plan by creating a question.

  FIG. 26 is an explanatory diagram showing an example of the automatic creation setting screen. The automatic creation setting screen d02 is a screen for performing necessary settings when automatically creating a plan. The automatic creation setting screen d02 includes an acquisition method setting d021, a period setting d022, a unit setting d023, and a creation button d024. In the acquisition method setting d021, a method for acquiring actual result data is set. The acquisition method setting d021 includes an upload link d0211 and an API (Application Programming Interface) cooperation link d0212. When the upload link d0211 is selected, a screen for setting a CSV (Comma-Separated Values) file to be uploaded is displayed, and the CSV file is uploaded by the screen. When the cooperation link d0212 is selected, a screen for setting API cooperation is displayed, and the setting for acquiring the result data from the other system 4 by API cooperation is performed. In the period number setting d022, the number of periods for planning is set. In unit setting d023, a unit for creating a plan is set. A plan is created by selecting the create button d024. The creation button d024 is inactive and cannot be selected until the method of acquiring the actual result data is set by the acquisition method setting d021.

  FIG. 27 is an explanatory diagram showing an example of a question answer screen. The question answer screen d03 is a screen for inputting an answer to a question presented when creating a question. The question answer screen d03 includes a first answer setting d031, a second answer setting d032, a third answer setting d033, and a next button d034. The first answer setting d031 is a pull-down menu for setting the answer to the first question. The second answer setting d032 includes a numerical value input box d0321 and a unit setting menu d0322. A numerical value for the answer is input in the numerical value input box d0321. The unit setting menu d0322 sets a unit for a numerical value. The third answer setting d033 is a pull-down menu for setting an answer to the third question. The next button d034 is a button for confirming the answer and moving to the next screen. The question displayed on the question answer screen d03 may be one question, two questions, or four or more questions. What is displayed as an answer setting is determined according to the question content. The content is stored in the logic DB 147.

  FIG. 28 is an explanatory diagram showing an example of the plan display screen. The plan display screen d04 is a screen for displaying the created business plan. The plan display screen d04 includes a plan value table d041, an enter button d042, and an edit button d043. The plan values created are displayed in the plan value table d041. The decision button d042 is a button for deciding the created plan as it is. The edit button d043 is a button for editing the created plan.

  FIG. 29 is an explanatory diagram showing an example of the editing method selection screen. The editing method selection screen d05 is a screen for selecting the editing method when editing the plan automatically created or the question created. The edit method selection screen d05 includes a direct edit button d051, a graph edit button d052, and a simulation edit button d053. When the user selects the direct edit button d051, the management server 1 executes the direct edit process. When the user selects the graph edit button d052, the management server 1 executes the graph edit process. When the user selects the simulation edit button d053, the management server 1 executes the simulation edit process.

  FIG. 30 is an explanatory diagram showing an example of the direct edit screen. The direct edit screen d06 is a screen for directly editing the created plan. A plan value table d061 and a decision button d062 are included. The planned value created in the planned value table d061 is displayed. It is possible to change the planned value by selecting the cell as in the spreadsheet software. The decision button d062 is a button for confirming the change result and deciding the plan.

  FIG. 31 is an explanatory diagram showing an example of a graph edit screen. The graph edit screen d07 is a screen for editing the created plan by a graph. The graph edit screen d07 includes a planned value table d071, a planned graph d072, and an enter button d073. The planned value created in the planned value table d071 is displayed. The plan graph d072 is a graph display of the plan value table d071. It is not necessary to graphically display all the subjects displayed in the plan value table d071. In FIG. 31, sales, selling, general and administrative expenses (selling and general administrative expenses), and operating profit are displayed in a graph. The planned value can be changed by dragging and dropping one point on the graph with a pointing device. Changes made by the graph are reflected in the planned values in the planned value table d071. The decision button d073 is a button for confirming the change result and deciding the plan.

  FIG. 32 is an explanatory diagram showing an example of the simulation edit screen. The simulation edit screen d08 is a screen for setting the created plan when the created plan is edited by simulation. The simulation edit screen d08 includes a plan value table d081, setting areas d082 to d084, and an enter button d085. The plan values created are displayed in the plan value table d081. Setting areas d082 to d084 are areas for setting simulation conditions. In the setting area d082, the sales amount can be set. In the setting area d083, SG & A expenses can be set. In the setting area d084, the variable cost ratio can be set. The enter button d085 is a button for confirming the setting and starting the recalculation of the planned value according to the simulation condition.

  As described above, the plan management system 100 has the following effects when creating a business plan. Since the plan is created using the actual data, it is possible to create a highly realistic plan. In addition, since it can be created by automatic creation or question creation, even a user who is not very familiar with management can create a plan. Further, since the created plan can be edited, the plan created by the plan management system 100 can be changed to a plan more in line with the user's intention. The plan can be changed intuitively by providing the graph editing function as an editing function. By providing the simulation editing function as the editing function, it is possible to finely adjust the plan, for example, to increase the sales by 5%.

  Next, the functions that the user can use during the planned operation in the plan management system will be described. FIG. 33 is a flowchart showing a procedure example of the planned operation process. The planned operation process is a process that provides functions related to the planned operation. The CPU 11 of the management server 1 transmits the menu screen to the user terminal 2 (step S201). The user selects the function to be executed from the menu screen. The user terminal 2 transmits the selected function to the management server 1. The CPU 11 acquires the selected function (step S202). The CPU 11 determines whether or not the selected function is predictive management (step S203). When the CPU 11 determines that the selected function is the prediction / actual management (YES in step S203), the CPU 11 performs the prediction / actual management process (step S204). When the CPU 11 determines that the selected function is not the actual management (NO in step S203), the CPU 11 determines whether the selected function is the report setting (step S205). When the CPU 11 determines that the selected function is the report setting (YES in step S205), it performs the report setting process (step S206). When the CPU 11 determines that the selected function is not the report setting (NO in step S205), the CPU 11 determines whether the selected function is the alert setting (step S207). When the CPU 11 determines that the selected function is alert setting (YES in step S207), it performs alert setting processing (step S208). When the CPU 11 determines that the selected function is not the alert setting (NO in step S207), the CPU 11 performs prediction / recreation processing (step S209). The CPU 11 ends the planned operation process.

  FIG. 34 is a flowchart showing an example of the procedure of the prediction / actual management process. The actual and actual management process is a process for setting the actual and actual management target and displaying the status of the management target. The CPU 11 of the management server 1 transmits the menu screen to the user terminal 2 (step S221). The CPU 21 of the user terminal 2 displays the menu screen on the display unit 26 (step S222). The user uses the input unit 25 to select a menu. The CPU 21 transmits the menu (selection menu) selected by the user to the management server 1 (step S223). The CPU 11 acquires the selection menu (step S224). The CPU 11 determines whether or not the selection menu is setting (step S225). When the CPU 11 determines that the selection menu is setting (YES in step S225), the setting screen is transmitted to the user terminal 2 (step S226). The CPU 21 of the user terminal 2 displays the setting screen on the display unit 26 (step S227). The user uses the input unit 25 to make settings. The CPU 21 transmits the setting information to the management server 1 (step S228). The CPU 11 of the management server 1 acquires the setting information (step S229). The CPU 11 stores the setting information (step S230). The setting information is stored in, for example, the prediction DB 148. When the CPU 11 determines that the selection menu is not set (NO in step S225), the CPU 11 acquires preliminary / actual data (step S231). The CPU 11 transmits a preliminary / realistic screen including the obtained preliminary / realistic data to the user terminal 2 (step 232). The CPU 11 returns the process to the calling source.

  FIG. 35 is a flow chart showing an example of the procedure of the report setting process. The report setting process is a process of setting the report created for the plan. The CPU 11 of the management server 1 transmits the menu screen to the user terminal 2 (step S241). The CPU 21 of the user terminal 2 displays the menu screen on the display unit 26 (step S242). The user uses the input unit 25 to select a menu. The CPU 21 transmits the menu selected by the user (selection menu) to the management server 1 (step S243). The CPU 11 acquires the selection menu (step S244). The CPU 11 determines whether or not the selection menu is the automatic creation (step S245). When the CPU 11 determines that the selection menu is not automatic creation (NO in step S245), the setting screen is transmitted to the user terminal 2 (step S246). The CPU 21 of the user terminal 2 displays the setting screen on the display unit 26 (step S247). The user uses the input unit 25 to set the report to be created and gives an instruction to send the report to the management server 1. The setting of the report is the selection of the subjects included in the standard report, etc. The CPU 21 transmits the setting information to the management server 1 (step S248). The CPU 11 of the management server 1 acquires the setting information (step S249). The CPU 11 stores the acquired setting information (step S250). When the CPU 11 determines that the selection menu is entrusted creation (YES in step S245), it stores that entrusted creation (an entrusted information) (step S251). The CPU 11 transmits the report setting screen to the user terminal 2 (step S252). The CPU 21 of the user terminal 2 displays the report setting screen on the display unit 26 (step S253). The user uses the input unit 25 to make report settings. The report settings include the report frequency, method (email, SNS, pop-up) and address. The CPU 21 transmits the report setting information to the management server 1 (step S254). The CPU 11 of the management server 1 acquires the report setting information (step S255). The CPU 11 stores the report setting information (step S256). The CPU 11 returns the process to the calling source. The setting information, the entrusted information, and the report setting information are stored in, for example, the report DB 149.

  FIG. 36 is a flowchart showing an example of the procedure of the alert setting process. The alert setting process is a process for setting an alert. The alert is to warn the user when the numerical value regarding the plan satisfies a predetermined condition set in advance. The CPU 11 of the management server 1 transmits the setting screen to the user terminal 2 (step S261). The CPU 21 of the user terminal 2 displays the setting screen on the display unit 26 (step S262). The user uses the input unit 25 to set an alert. The alert setting is, for example, to send an alert when the difference between the planned value and the actual value of sales is 10% or more of the planned value. The CPU 21 transmits the alert setting to the management server 1 (step S263). The CPU 11 of the management server 1 acquires the alert setting (step S264). The CPU 11 stores the alert setting in the alert DB 14A (step S265). The CPU 11 returns the process to the calling source.

  FIG. 37 is a flowchart showing an example of the procedure of batch processing. The batch process is a process for correcting the data relating to actual results. The CPU 11 of the management server 1 performs a preliminary / realistic creation process (step S271). The CPU 11 performs a report creation process (step S272). The CPU 11 performs alert creation processing (step 273). Batch processing is executed daily, for example.

  FIG. 38 is a flow chart showing an example of the procedure of the preliminary and actual creation processing. The actual / realistic creation processing collects actual / realistic data regarding numerical values to be managed and calculates the difference between actual and actual. The CPU 11 of the management server 1 acquires the actual / real setting from the actual / real DB 148 (step S281). The CPU 11 selects one unprocessed prediction / actual setting (step S282). The CPU 11 determines whether it is the execution timing (step S283). For example, when the actual and actual management is set monthly in the actual and actual setting, and if the actual and actual for the previous month has not been created, it is the execution timing. If the time of execution is after the closing time of the closing day, it is the execution timing. If not, it is not the execution timing. When the CPU 11 determines that it is not the execution timing (NO in step S283), the process proceeds to step S287. When it is determined that it is the execution timing (YES in step S283), the CPU 11 acquires the plan data and the actual result data (step S284). If the target of the actual management is sales, the CPU 11 acquires plan data and actual data from the accounting DB 142. The CPU 11 creates preliminary and actual data (step S285). For example, the actual data is a set of plan data, actual data, and difference data between them. The CPU 11 stores the actual data in the actual DB 148 (step S286). The CPU 11 determines whether or not there is an unprocessed preliminary / actual setting (step S287). When the CPU 11 determines that there is an unprocessed actual / real setting (YES in step S287), the process returns to step S282, and the process for the unprocessed actual / real setting is performed. When the CPU 11 determines that there is no unprocessed prediction / actual setting (NO in step S287), the prediction / actual creation process is terminated and the process is returned to the calling source.

  FIG. 39 is a flow chart showing an example of the procedure of the report creation processing. The report creation process is a process of creating and transmitting a report according to the report settings. The CPU 11 of the management server 1 acquires the report settings from the report DB 149 (step S291). The CPU 11 selects one unprocessed report setting (step S292). The CPU 11 determines whether it is the reporting timing (step S293). For example, in the report setting, if the report is set to be created monthly, and the report for the previous month has not been created, it is the report timing. If the execution time is after the closing time of the closing day, it is the reporting timing. If not, it is not the reporting timing. When the CPU 11 determines that it is not the report timing (NO in step S293), the process proceeds to step S297. When the CPU 11 determines that it is the reporting timing (YES in step S293), the CPU 11 determines whether the creation method is entrusted creation (step S294). When the CPU 11 determines that the creating method is not the automatic creation (NO in step S294), the standard report is created according to the report setting (step S295). When the CPU 11 determines that the creating method is the automatic creation (YES in step S294), it automatically creates the report (step S298). The CPU 11 transmits the report (step S296). The CPU 11 transmits the report by a method according to the report setting of the process target user stored in the report DB 149. The CPU 11 determines whether or not there is an unprocessed report setting (step S297). When determining that there is an unprocessed report setting (YES in step S297), the CPU 11 returns the process to step S292, and performs a process for the unprocessed report setting. When the CPU 11 determines that there is no unprocessed report setting (NO in step S297), the report creating process is terminated and the process is returned to the calling source.

  In creating the standard report, the CPU 11 obtains the plan data and the actual data of the target item each time the period elapses according to the target item selected from the plurality of items (subjects) and the period (for example, one month). get. The CPU 11 outputs a report about the target item. At that time, the CPU 11 uses the actual data and the like created in the actual creation process.

  In the automatic creation, the CPU 11 calculates the degree of deviation (for example, difference value) from the actual data for each item (subject) of the plan data. The CPU 11 determines the priority order for each item based on the calculated degree of deviation for each item. CPU11 produces | generates the report which laid out each item based on the determined priority.

  FIG. 40 is an explanatory diagram showing an example of the automatic report creation logic. The example shown in FIG. 40 is an example in which a report is created from past performance data instead of the actual data. FIG. 40 shows an example in which the factors that increase operating profit are analyzed. The subject to be analyzed may be designated by the user in advance, or may be determined by the CPU 11 of the management server 1. In the example of FIG. 40, sales, cost of sales, variable cost ratio, gross profit, selling expenses, general and administrative expenses (SG & A expenses), and operating profit are increasing as subjects. The CPU 11 obtains the difference between the actual values for 2016 and 2017 for each subject. Further, the CPU 11 calculates the increase / decrease rate. The CPU 11 compares the increase / decrease rates of the respective subjects, and selects the subject having the largest absolute value as the subject of analysis. In the example of FIG. 40, the largest absolute value of the increase / decrease rate is the operating profit, so the CPU 11 targets the operating profit for analysis. Next, the CPU 11 ranks the items related to operating profit in the order of the absolute value of the increase / decrease rate. Here, gross profit is first, sales are second, variable cost ratio is third, cost of sales is fourth, and SG & A expenses are fifth. This ranking is the order of the factors that affected the increase in operating profit. The CPU 11 lays out the display of each subject in the output report, for example, a graph so that the user can see it in order of rank. This allows the user to intuitively understand the ranking of the factors that have affected the increase in operating profit.

  Here, the actual data is taken as an example, but the same processing can be performed with the actual data. For example, when the operating profit for a certain month exceeds the planned value by a large amount, the difference between the planned value and the actual value for each item of P / L is calculated and converted into an increase / decrease rate with respect to the planned value. Then, the priority may be obtained based on the magnitude of the increase / decrease rate.

  FIG. 41 is a flowchart showing an example of the procedure of the alert creating process. The alert creating process is a process of creating and transmitting an alert when the actual / real management target satisfies the alert condition according to the alert setting. The CPU 11 of the management server 1 acquires alert settings from the alert DB 14A (step S311). The CPU 11 selects one unprocessed alert setting (step S312). The CPU 11 determines whether or not it is an alert timing (step S313). For example, when an alert about sales is set, and when the actual / real data can be collected daily, the alert timing is every day. If the actual data can only be collected monthly, the alert will be issued only on the 1st of the month. When the CPU 11 determines that it is not an alert timing (NO in step S313), the process proceeds to step S318. When the CPU 11 determines that it is an alert timing (YES in step S313), the evaluation index is acquired (step S314). The evaluation index is an index used to determine whether or not to send an alert. For example, the actual value of sales. The CPU 11 determines whether or not the condition of the alert is satisfied (step S315). When the CPU 11 determines that the alert condition is not satisfied (NO in step S315), the process proceeds to step S318. When the CPU 11 determines that the condition of the alert is satisfied (YES in step S315), the CPU 11 creates the alert (step S316). The CPU 11 transmits the created alert (step S317). The CPU 11 sends an alert by a user-specified method included in the alert setting. The CPU 11 determines whether or not there is an unprocessed alert setting (step S318). When determining that there is an unprocessed alert setting (YES in step S318), the CPU 11 returns the process to step S312, and performs a process for the unprocessed alert setting. When the CPU 11 determines that there is no unprocessed alert setting (NO in step S318), the alert creating process ends and the process returns to the calling source.

  FIG. 42 is a flowchart showing a procedure example of the prediction / re-creation processing. The forecasting / recreating process is a process of predicting the fluctuation of the numerical values included in the plan and recreating the plan. The CPU 11 of the management server 1 transmits the menu screen to the user terminal 2 (step S331). The CPU 21 of the user terminal 2 displays the menu screen on the display unit 26 (step S332). The user uses the input unit 25 to select a menu to be executed. The CPU 21 sends the menu selected by the user (selection menu) to the management server 1 (step S333). The CPU 11 of the management server 1 acquires the selection menu (step S334). The CPU 11 determines whether or not the selection menu is recreated (step S335). When the CPU 11 determines that the selection menu is re-creation (YES in step S335), the method selection screen is transmitted to the user terminal 2 (step S336). The CPU 21 of the user terminal 2 displays the method selection screen on the display unit 26 (step S337). The user uses the input unit 25 to select a recreating method. The CPU 21 sends the recreating method to the management server 1 (step S338). The CPU 11 of the management server 1 acquires the recreating method (step S339). The CPU 11 recreates the plan according to the acquired method (step S340). The CPU 11 stores the recreated plan in the accounting DB 142 or the like (step S341). The CPU 11 moves to step S5 of FIG. The user can edit the plan recreated by the management server 1 as in the case of creating the plan. When the CPU 11 determines that the selection menu is not recreated (NO in step S335), it makes a prediction (step S343). For example, from the sales results of each month of the first quarter, the sales of each month after the second quarter are predicted. The CPU 11 transmits the prediction result to the user terminal 2. The CPU 11 ends the prediction / re-creation processing and returns the processing to the calling source.

  FIG. 43 is an explanatory diagram showing an example of the prediction logic. FIG. 43 shows an example of predicting the sales after April 2018 from the actual sales of January to March 2018. The unit of sales is, for example, 10,000 yen. The upper part of FIG. 43 shows a method of obtaining the seasonal variation index for each month from the sales amount for each month in the three years before forecast (2015 to 2017). The seasonal variation index shows the percentage of the monthly sales to the annual sales as a percentage. The lower part of FIG. 43 predicts the sales from each month to the annual sales based on the sales from January to March and the seasonal variation index from January to March. From January sales of 1,300 and January seasonal fluctuation index of 9.3%, the forecast sales are 13,919. The annual sales are 13,283 from the February sales, and the annual sales are 14,311 from the March sales. Further, the predicted average value of annual sales is used as the final predicted value of annual sales. Here, the average value 13,838 of the values of 13,919, 13,283, 14,311 is the final predicted annual sales. From this value and the seasonal fluctuation index for each month after April, the forecast value of sales for each month is obtained. The middle part of FIG. 43 shows the final prediction result. Cost of sales, SG & A expenses, ordinary income, etc. can be predicted by the same method as sales. Further, by forecasting B / S, it is possible to forecast the tax amount and cash balance at the end of the period. Based on these forecast figures, it is possible to judge changes in the management plan, such as making capital investments toward the end of the fiscal year and refraining from initial planned investments.

  Next, the screen displayed on the user terminal 2 in the above-described processing will be described. FIG. 44 is an explanatory diagram showing an example of the operation menu screen. The operation menu screen d09 is a screen for selecting a function to be executed among the functions related to the total planning operation. The operation menu screen d09 includes a forecast / actual management button d091, a report setting button d092, an alert setting button d093, and a forecast / recreate button d094. The actual and actual management button d091 is a button for executing the actual and actual management process. The report setting button d092 is a button for executing a report setting process. The alert setting button d093 is a button for executing an alert setting process. The prediction / re-creation button d094 is a button for executing the prediction / re-creation processing.

  FIG. 45 is an explanatory diagram showing an example of the prediction / actual management menu screen. The actual and actual management menu screen d10 is a screen for selecting a function to be executed among the functions related to actual and actual. The actual management menu screen d10 includes a setting button d101 and a reference button d102. The setting button d101 is a button for setting the target of preliminary and actual management. The reference button d102 is a button for referring to the actual situation.

  FIG. 46 is an explanatory diagram showing an example of a preliminary / actual management setting screen. The actual and actual management setting screen d11 is a screen for setting the actual and actual management target. The actual management setting screen d11 includes a first setting area d111, a second setting area d112, an add button d113, and an enter button d114. The first setting area d111 includes an index selection menu d1111 and a timing menu d1112. The second setting area d112 includes an index selection menu d1121 and a timing menu d1122. The index selection menu d1111 and the index selection menu d1121 are menus for selecting an index to be the target of predictive management. The timing menu d1112 and the timing menu d1122 are menus for setting units to be managed. The add button d113 is used when three or more targets of the actual management are set. When the add button d113 is operated, the setting area is additionally displayed. The decision button d114 is a button for deciding the setting content.

  FIG. 47 is an explanatory diagram showing an example of a prediction / actual situation display screen. The actual / actual situation display screen d12 is a screen for displaying the actual situation. The preliminary and actual situation display screen d12 includes a selection menu d121, a management content d122, a preliminary and actual table d123, and a close button d124. The selection menu d121 is a menu for selecting a management target to be displayed. The management content d122 displays the content of the actual management. The forecast table d123 shows the forecast situation in a table. The close button d124 closes the prediction / actual situation display screen d12. When a plurality of management targets have the same unit, they may be displayed collectively. For example, when the sales amount is managed on a monthly basis and the operating profit is also managed on a monthly basis, it may be displayed not on an individual basis but on one forecast table d123.

  FIG. 48 is an explanatory diagram showing an example of the report selection screen. The report selection screen d13 is a screen for selecting a report to be created. The report selection screen d13 includes a standard report creation button d131, an automatic report button d132, and a close button d133. The standard report creation button d131 is a button for making settings for creating a standard report. The automatic report button d132 is a button for automatically generating a report. The close button d133 is a button for closing the report selection screen.

  FIG. 49 is an explanatory diagram showing an example of the standard report setting screen. The standard report setting screen d14 is a screen for setting the contents of the standard report. The standard report setting screen d14 includes an item setting table d141 and an enter button d142. The item setting table d141 has a display similar to that of the spreadsheet software. The user uses the input unit 25 to set the subject etc. to be displayed in the report in the item setting table d141. Select the input content from the pull-down menu. Or, the user types. In this case, the pattern matching of the input character string is performed, and the input of the character string other than the previously registered character string is prohibited. The decision button d142 is a button for deciding the setting content. In the example shown in FIG. 49, the monthly plan value, the actual value, and the difference value between the planned value and the actual value are displayed for each of the sales amount and the SG & A cost. , The cumulative plan value, cumulative actual value, and cumulative difference value up to this month are set in a table.

  FIG. 50 is an explanatory diagram showing an example of the report setting screen. The report setting screen d15 is a screen for setting a report transmission destination and the like. The report setting screen d15 includes a first setting area d151, a second setting area d152, an add button d153, and an enter button d154. The first setting area d151 and the second setting area d152 are areas for setting a report destination. The first setting area d151 includes a frequency menu d1511, a method menu d1512, and an address box d1513. The frequency menu d1511 sets the frequency of reporting. The method menu d1512 sets the reporting method. The address of the report is entered in the address box d1513. If the reporting method is e-mail, enter the e-mail address, and if it is a message service, enter the ID. The second setting area d152 includes frequency menus d1521, d1522, a method menu d1523, and an address box d1524. The frequency menus d1521 and d1522 enter the frequency of reporting. Here, a case is shown in which two parameters are input as the frequency every day at 5 pm. The method menu d1523 and the address box d1524 are similar to the method menu d1512 and the address box d1513, respectively. The add button d153 is used when three or more report destinations are set. When the add button d153 is operated, the setting area is additionally displayed. The decision button d154 is a button for deciding the setting content.

  FIG. 51 is an explanatory diagram showing an example of an automatic report. As shown in FIG. 51, a graph of management-related indexes is displayed in order of priority. If each graph shown in FIG. 51 is a screen display, one graph is displayed on one screen. For paper documents, one graph is printed on each page. As a result, the user can appropriately grasp the index that has affected the management.

  FIG. 52 is an explanatory diagram showing an example of the alert setting screen. The alert setting screen d16 is a screen for setting alert conditions. The alert setting screen d16 includes an index menu d161, a threshold box d162, a unit menu d163, a method menu d164, an address box d165, an add button d166, and an enter button d167. An index to be monitored is set in the index menu d161. In the threshold box d162, a reference threshold value is input when determining whether to generate an alert. A unit of threshold is selected in the unit menu d163. The method menu d164 stores an alert notification method. In the address box d165, an alert notification destination is entered. The add button d166 is a button used when adding an alert condition. The decision button d167 is a button for deciding the alert setting.

  FIG. 53 is an explanatory diagram showing an example of the prediction / re-creation menu screen. The prediction / re-creation menu screen d17 is a screen for selecting execution of either prediction or re-creation. The prediction / re-creation menu screen d17 includes a re-creation button d171, a prediction button d172, and a close button d173. The recreating button d171 is a button for recreating a plan. The forecast button d172 is a button for forecasting sales and the like. The close button d173 is a button for closing the prediction / recreate menu screen d17.

  FIG. 54 is an explanatory diagram showing an example of a reproduction method selection screen. The recreating method selection screen d18 is a screen for selecting a method for recreating a plan. The recreating method selection screen d18 includes an automatic creating button d181, a direct input creating button d182, a graph edit creating button d183, a simulation button creating d184, and a close button d185. The automatic creation button d181 is a button for recreating by automatic creation. The direct input creation button d182 is a button for recreating by direct input. The graph edit creation button d183 is a button for recreating by graph operation. The simulation button creation d184 is a button for recreating by simulation. Whichever method is used, re-creation is performed by the same process as plan editing. The close button d185 is a button for closing the recreating method selection screen d18.

  As described above, the plan management system 100 has the following effects during the operation of the business plan. Since it is possible to create a report of actual and actual management, it is possible to quickly take breakthrough measures when the actual and actual deviations are large. If there is no effective breakthrough, it is possible to recreate the plan. In addition, it is possible to receive an alert when the difference between the actual and the actual is large. In particular, if an alert is set for an index that has a great influence on the business situation and the creation cycle is set shorter than the report creation cycle, it is possible to respond more quickly. In addition, since it is possible to predict the situation at the end of the term in the middle of the term, it is possible to take additional measures with an eye on the situation at the end of the term.

  On the actual / real situation display screen d12, an advice message according to the actual situation or an advertisement based on the advice content may be displayed. Similarly, the alert message may include an advice message or advertisement.

  FIG. 55 is an explanatory diagram showing an example of the advice DB. The advice DB 14B is stored in the large-capacity storage unit 14 of the management server 1, for example. The advice DB 14B stores the advice displayed on the prediction / actual situation display screen d12 and the advice added to the alert. The advice DB 14B includes an advice ID string, an evaluation index string, a condition string, and an advice string. The advice ID column stores the advice ID that can uniquely identify the advice. The evaluation index column stores an index (subject name) corresponding to the advice. The condition column stores conditions for adding advice. The advice column stores the advice to be displayed or transmitted.

  FIG. 56 is an explanatory diagram showing an example of the advertisement DB. The advertisement DB 14C is stored in, for example, the large capacity storage unit 14 of the management server 1. The advertisement DB 14C stores the advertisement to be displayed or transmitted together with the advice. The advertisement DB 14C includes an advertisement ID column, an advice ID column, a text column, and a link column. The advertisement ID column stores an advertisement ID that can uniquely identify the advertisement. The advice ID column stores the corresponding advice ID. The text string stores the text of the advertisement. The link string stores hyperlinks embedded in text. The hyperlink is, for example, a URL (Uniform Resource Locator) of a home page that describes in detail the advertised product or service.

  FIG. 57 is a flow chart showing another example of the procedure of the preliminary and actual creation processing and the alert creation processing. FIG. 57A shows another example of the procedure of the preliminary and actual creation processing. FIG. 57B shows another example of the procedure of the alert creation processing. 57A and 57B show a part of the flowchart. As shown in FIG. 57A, in the preliminary / actual creation process, a process is added to the procedure shown in FIG. 38 and a part of the process is changed. In the actual / realistic creation processing, the CPU 11 of the management server 1 selects the advice (step S288) after executing the actual / realistic data creation (step S285). Specifically, the CPU 11 compares the prepared actual data with the evaluation index sequence and the situation sequence of the advice DB 14B, and the evaluation data that matches the conditions defined in the evaluation index sequence and the situation sequence is included in the actual data. It is determined. When determining that there is an evaluation index that matches the condition, the CPU 11 selects the corresponding advice. The CPU 11 selects an advertisement corresponding to the selected advice (step S289). Specifically, the CPU 11 searches the advertisement DB 14C using the advice ID of the selected advice as a search key. When a plurality of advertisements hit the search, they are randomly selected, for example. The CPU 11 stores the selected advice and advertisement together with the created prediction data (step S286). The CPU 11 executes step S287 and the subsequent steps.

  As shown in FIG. 57B, in the alert creation processing, processing is added to the procedure shown in FIG. 41 and some processing is changed. In the alert creating process, the CPU 11 of the management server 1 selects the advice (step S319) after executing the alert creating (step S316). Specifically, the CPU 11 determines whether the alert creation conditions satisfy the conditions defined in the evaluation index column and the situation column of the advice DB 14B. When determining that the condition is met, the CPU 11 selects the advice corresponding to the condition. The CPU 11 selects an advertisement corresponding to the selected advice (step S320). Step S320 is the same as step S89 described above. The selected advice and advertisement are transmitted together with the generated alert (step S317). The CPU 11 executes step S318 and subsequent steps.

  In the case of the Omakase Report, the advice may be selected based on the subject to be analyzed or the subject that has influenced the subject to be analyzed. Then, the advertisement is selected based on the selected advice.

  FIG. 58 is an explanatory diagram showing an example of the cooperation setting screen. The cooperation setting screen d19 is a screen for making settings for data cooperation with the other system 4. The cooperation setting screen d119 includes an accounting selection menu d191, a POS selection menu d192, an attendance selection menu d193, a reservation selection menu d194, a review site input field d195, and a cooperation button d196. The accounting selection menu d191 is a pull-down menu and selects an accounting system to be linked. The POS selection menu d192 is a pull-down menu and selects a POS system to be linked. The attendance selection menu d193 is a pull-down menu for selecting an attendance system to be linked. The reservation selection menu d194 is a pull-down menu for selecting a reservation system to be linked. In the word-of-mouth site input field d195, enter the URL of the word-of-mouth site to be linked. When the mouse is clicked on the cooperation button d196, the cooperation setting is stored.

  The management server 1 acquires user data held by various cloud services at any time by API cooperation according to the cooperation setting screen d19. The acquired data is stored in the database of the mass storage unit 14. Various cloud services include a reservation management service, a POS (Point Of Sale) service, an attendance management service, an accounting service, a payroll processing service, an order management service, and the like. It is desirable that the cloud service transmits its contents to the management server 1 every time the user data is updated. If that is difficult, the management server 1 polls the cloud service.

  FIG. 59 is an explanatory diagram showing the multi-CSV upload function. The CSV here is a collection of a plurality of subjects (indexes) in one file. Each line is one subject. FIG. 59A is an example of converting CSVs in which the order of values in each row is the same but the number of rows in the subject is different. In both the CSV format 561 of company X and the CSV format 562 of its own company, each line is “subject, amount, date and time, remarks”. However, the second line of company X has a "sales discount", while the second line of its company has a "cost of sales". “Cost of sales” is the fourth line for company X. Similarly, the “labor cost” is the 9th line for Company X and the 3rd line for the company. In such a case, the conversion logic 563 is prepared as data for conversion and stored in the mass storage unit 14. Like the conversion logic 563, by preparing the conversion logic 563 in which the fourth line of the X company corresponds to the second line of the company and the ninth line of the X company corresponds to the third line of the company, the X company It is possible to convert the CSV file of to the CSV file of the company. If the CSV format of the company and the conversion logic 563 are registered in the management server 1, for example, the plan data exported from the sales system of the company and the actual data exported from the POS system of the company X are uploaded to the management server 1. However, it is possible to manage sales forecasts.

  FIG. 59B is an example of converting CSV in which the number of rows in which subjects appear is the same, but the arrangement order of values is different in each row. Both the Y company's CSV format 564 and its company's CSV format 565 have the same row arrangement order, such as the first line is “sales” and the second line is “cost of sales”. However, each row of company Y has a sequence of "sales, date and time, amount of money, remarks", whereas each row of its company has a sequence of "sales, amount of money, date and time and remarks". In such a case, the conversion logic 566 is prepared as data for conversion and stored in the mass storage unit 14. Like the conversion logic 566, by preparing the conversion logic 566 that the second value (column B) in each row of Y company corresponds to the third value (column C) in each row of the company Y, It is possible to convert the CSV file of the company to the CSV file of the company. If the CSV format of the company and the conversion logic 566 are registered in the management server 1, for example, the plan data exported from the procurement system of the company and the actual data exported from the purchasing system of the company Y are uploaded to the management server 1. However, it becomes possible to manage the actual cost of sales.

Embodiment 2
This embodiment relates to an embodiment in which a plan is created by artificial intelligence. In this embodiment, a plan is created based on past performance data. FIG. 60 is an explanatory diagram showing an example of the collection item DB. The collection item DB 14D stores data items to be collected when creating a plan. The collection item DB 14D includes a business type column, a prediction item sequence, and a collection item sequence. The business type column stores the business types. The prediction item sequence stores data items for which prediction is performed. The collection item string stores data items to be collected. In the example shown in FIG. 60, it is defined that the number of stores, the number of seats, and the average price should be collected in order to plan the sales amount of the restaurant. The number of seats is the total number of seats in all stores. The average price is the average price of the main menu. For example, at a ramen shop, it is the average price of all types of ramen.

  FIG. 61 is an explanatory diagram showing an example of the sales DB. The sales DB 14E is a collection of past sales of restaurants based on the collection item DB 14D. The sales DB 14E includes a year column, a sales column, a store number column, a seat number column, and an average price column. The year column stores the fiscal year. The sales column stores the sales. The number of stores column stores the number of stores. The number of seats column stores the total number of seats. The average price column stores the average price. The number of stores, the number of seats, and the average price shall be values at predetermined points such as the beginning of the year, the first half of the year, and the end of the year. In addition, the average value at the end of every month may be used in consideration of the case of fluctuations during the year.

  FIG. 62 is an explanatory diagram related to the sales forecast model. The outline of the present embodiment will be described below. The management server 1 performs machine learning for learning the sales amount for each year to generate the sales amount prediction model 14F. For example, the sales forecast model 14F is a neural network related to LSTM (Long-Short Term Memory). The LSTM is a type of RNN (Recurrence Neural Network), and is a neural network that inputs time-series data before the prediction target time point and outputs a prediction value at the target time point.

  The LSTM has an input layer, an intermediate layer, and an output layer. The input layer has a plurality of neurons that respectively accept the input of parameters at each time point along a time series. The output layer has a neuron that outputs a predicted value of sales. The middle layer has neurons for calculating a prediction value from the input value to each neuron of the input layer. The neuron in the middle layer is called an LSTM Block, and by performing the calculation on the input value at the next time using the calculation result in the middle layer on the input value at the past time, the time series data up to the latest time can be used. Calculate the value at the next time.

  The configuration of the LSTM shown in FIG. 62 is an example, and the present embodiment is not limited to this. For example, the intermediate layer is not limited to one layer and may be two or more layers. Further, the number of neurons in the output layer is not limited to one. The sales forecast model 14F is not limited to the LSTM and may be a neural network other than the LSTM.

  The management server 1 uses the time-series data of the actual values of various parameters stored in the sales DB 14E as teacher data to generate the sales forecast model 14F. Specifically, the management server 1 sequentially inputs the actual values of the year, the sales amount, the number of stores, the number of seats, and the average price at each time point to the corresponding neurons of the input layer in a time series, and at a certain time point in the past. Calculate the forecast value of sales in. The management server 1 compares the actual sales amount at the target time as the correct value with the predicted value, and constructs the sales amount prediction model 14F.

  As described above, the management server 1 uses the time series data of the sales to be predicted and the time series data of the parameters other than the sales such as the year, the number of stores, the number of seats, and the average price in the sales prediction model 14F. Enter and learn.

  The management server 1 sequentially inputs the sales amount and other parameters to each neuron. The management server 1 performs the calculation in the middle tier, and finally outputs the predicted value of the sales amount at the target time from the neuron in the output layer. The management server 1 compares the actual sales amount at the target time with the predicted value as the correct value, and optimizes the parameter used for the calculation in the middle tier so that the predicted value approximates the correct value. The parameter is, for example, a weight between neurons (coupling coefficient), a coefficient of an activation function, or the like. The method of optimizing the parameters is not particularly limited, but the management server 1 optimizes various parameters using the error back propagation method, for example.

  Through the above-described processing, the management server 1 generates the sales forecast model 14F with the sales and other parameters. The management server 1 predicts sales by using the generated sales prediction model 14F.

  First, the management server 1 acquires the sales time series data, the year, the number of stores, the number of seats, and the average price from the sales DB 14E. In this case, the management server 1 is based on the number of stores, the number of seats, and the average price as well as the actual value at each time point in the past, and the number of stores and the number of seats based on the opening plan of the forecast year and the menu strategy of the forecast year. Get the average price.

  The management server 1 sequentially inputs the sales for each year, which constitutes the time-series data, the year, the number of stores, the number of seats, and the average price to the sales prediction model (learned model) 14F, as in the case of learning. That is, the management server 1 inputs the sales amount of each past year into each corresponding neuron of the input layer, and inputs the year, the number of stores, the number of seats, and the average price to each corresponding neuron. The management server 1 sequentially inputs the year, the number of stores, the number of seats, and the average price in time series, and finally inputs not only the actual value but also the predicted value to the sales forecast model 14F. Thereby, the management server 1 acquires the predicted value of the sales amount in the predicted year as the output from the output layer. As described above, the management server 1 inputs the sales amount and other parameters and makes a prediction.

  The above is a forecast of sales at restaurants. For industries other than restaurants, data items to be referred to are different, and prediction can be made in the same manner.

  Similarly to the sales amount, the forecast model is generated for the cost of sales, and the SG & A expenses and the general administrative expenses by learning using the past performance data as teacher data. The generated forecast model is used to forecast the cost of sales for the forecast year, as well as SG & A expenses. As described above, it is possible to generate the planned value of the main index of the profit and loss plan by the prediction by the artificial intelligence.

  FIG. 63 is a flowchart showing an example of the procedure of learning / prediction processing. The CPU 11 of the management server 1 acquires the business category of the user who is the prediction target (step S351). The business type of the user is stored in, for example, the user DB 141. The CPU 11 selects an item to be predicted, for example, sales amount (step S352). CPU11 produces | generates the prediction model regarding the selected item (step S353). The CPU 11 makes a prediction using the generated prediction model (step S354). The CPU 11 determines whether or not there is another unprocessed item (step S355). When the CPU 11 determines that there is another unprocessed item (YES in step S355), the process returns to step S352, and the process for the unprocessed item is performed. When the CPU 11 determines that there is no other unprocessed item (NO in step S355), it outputs the predicted value (step S356) and ends the process.

  FIG. 64 is a flowchart showing an example of the procedure of a prediction model generation process. The CPU 11 of the management server 1 acquires the teacher data used for generating the prediction model (step S361). For example, the CPU 11 reads, from the sales DB 14E, data indicating the actual value of sales in time series and data indicating the actual value of the year, the number of stores, the number of seats, and the average price in time series, and uses them as teacher data.

  The CPU 11 uses the teacher data to output the prediction value of the selected item when the time series data of the item selected in step S352 (hereinafter referred to as the selection item) and the time series data of the collection item such as the year are input. Is generated (step S362). The collection item is an item stored in the collection data string of the collection item DB 14D in association with the type of business and the selection item. When learning the selection item at a certain time point in the past, the CPU 11 inputs the actual value of the selection item at each time point included in the time-series data and the actual value of the collection item into the prediction model according to the time series, and at the target time point. Calculate the predicted value of the selected item. In this case, the CPU 11 compares the predicted value with the actual value of the selection item at the actual target time point, optimizes the weights of the intermediate layer and the like so that they approximate each other, and generates the prediction model. The model is stored in the mass storage unit 14 (step S363), and the process is returned to the calling source.

  FIG. 65 is a flowchart showing a procedure example of the prediction process. The CPU 11 acquires time series data of each parameter input to the prediction model (step S371). The CPU 11 selects a prediction model according to the selection item (step S372). The CPU 11 sequentially inputs the actual value of the selection item at each time point included in the time-series data and the actual value or predicted value of the collection item to the selected prediction model in time series (step S373). The CPU 11 acquires the predicted value of the selected item at the target time as an output from the prediction model (step S374). The CPU 11 stores the prediction result of the selection item output from the prediction model in the temporary area provided in the RAM 13 or the large-capacity storage unit 14 in association with the identification information (for example, item name) of the selection item (step S375). Return the process to the caller.

  In the above description, the generation of the prediction model (machine learning) and the prediction are performed by the management server 1, which is the same computer, but, for example, the management server 1 generates the prediction model and the generated prediction model is used by the user. After installing on the terminal 2, the user terminal 2 may perform prediction using the prediction model. That is, separate computers may execute the machine learning process and the process using the product of the machine learning.

  The above description is based on the premise that there is one industry for each user. For users who are engaged in business in multiple industries, a prediction model is generated and predicted for each industry. The predicted values for each industry are added together to make the predicted value for the entire user. Further, although the sales amount is predicted, values such as SG & A expenses (SG & A expenses and general administrative expenses) and variable cost ratios can also be predicted by the same method as described above.

  According to the present embodiment, it is possible to create a plan with a predicted value based on artificial intelligence using time series data as a planned value.

Embodiment 3
This embodiment relates to another embodiment in which a plan is created by artificial intelligence. In the second embodiment, time series data is used. In the present embodiment, data relating to the business strategy and business environment is used instead of the time series data. Here, the retail business is taken as an example. This is an example in which the sales amount of each store is predicted using artificial intelligence and the predicted sales amount is used as the planned value.

  FIG. 66 is an explanatory diagram showing an example of the sales DB. The sales DB 14F stores the annual sales (main items) of each store in association with variables (explanation items) that are considered to be correlated with the sales. The sales DB 14F includes a store ID column, a location information column, a location column, a nearest station column, a sales floor area column, an employee number column, an average price column, a parking lot number column, a station passenger number column, a vehicle traffic column, a 5 km population column, and Including sales column. The store ID column stores a store ID that can uniquely identify the store. The position information string stores the position of the store. The position is, for example, the latitude and longitude. The location column stores the location conditions of the store. For example, if the store is located within a 5-minute walk from a public transportation station, the vicinity of the station is stored. If the store is located in the suburbs, remember the suburbs. If the store is in a shopping center, the SC is stored. The nearest station line stores the nearest station of the store. The sales floor area column stores the total floor area of the sales floor portion of the store. The number of employees column stores the number of employees in the store. The average price column stores the average selling price of the products handled in the store. The parking lot number column stores the possible number of parking lots in the parking lot provided by the store for customers. When shared with other companies' stores, such as a shopping center, adjustments may be made by multiplying by a coefficient according to the number of stores in which the tenant is located. The station passenger number column stores the number of passengers a day at the nearest station. Weekdays and weekends and holidays may be stored separately. The vehicle traffic volume column stores the traffic volume per day on the main road used to access the store. The 5 km population column stores the population within a 5 km area centered on stores. The sales column stores the annual sales of the store.

  FIG. 67 is an explanatory diagram regarding the generation process of the prediction model. In FIG. 67, the process of performing machine learning to generate the prediction model 14G is conceptually illustrated.

  The management server 1 generates a neural network as the prediction model 14G. The back propagation method is used for learning the prediction model 14G. The back-propagation method gives teacher data in which an input value and a desired output value (correct value) corresponding to the input value are combined, and each node is combined so as to minimize the square error between the output value of the neural network and the correct value. Is a learning method to correct. The backpropagation method is also called an error backpropagation method. In the prediction model 14G, the position of the store, the location, the nearest station, ... The population of 5 km are input, and the annual sales amount is set as the correct value. Then, learning about all stores is performed. The learning may use not only data for a single year but data for multiple years.

  Next, the prediction of the sales amount using the learned prediction model (learned model) 14G will be described. FIG. 68 is a flowchart showing an example of the procedure of the prediction process. The CPU 11 of the management server 1 acquires the data of the existing store (step S381). The data to be acquired is the population of 5 km from the store ID and the number of employees among the items included in the sales DB 14F. If the number of employees or the average price changes as a strategy for the forecast year, update the value according to the strategy. If the number of parking lots, the number of passengers at the station, the traffic volume of the vehicle, and the population of 5 km are not the latest data, they are updated to the latest data. Alternatively, if there is a forecast value for the forecast year, it may be updated to that value. The CPU 11 acquires the data of the store scheduled to open (step S382). The data items to be acquired are the same as in the existing store. The CPU 11 selects a store to be processed (step S383). The CPU 11 inputs the data of the selected store into the prediction model 14G (step S384). The CPU 11 acquires the sales forecast value output by the prediction model 14G (step S385). The CPU 11 stores the predicted value in the temporary storage area (step S386). The temporary storage area is provided in the RAM 13 or the large-capacity storage unit 14. The CPU 11 determines whether there is an unprocessed store (step S387). When determining that there is an unprocessed store (YES in step S387), the CPU 11 returns the process to step S383, and performs the process for the unprocessed store. The processing from step S383 to step S386 is the same processing at the existing store and the store scheduled to open. When the CPU 11 determines that there is no unprocessed store (NO in step S387), the CPU 11 accumulates the predicted value of each store stored in the temporary storage area (step S388). The CPU 11 outputs the integrated value as a company-wide sales forecast value (step S389), and ends the process. Since the stores scheduled to leave the store and the stores scheduled to open stores are not in operation throughout the year, the estimated values obtained in step S385 are multiplied by a value obtained by multiplying the estimated values by a coefficient according to the business period. Good.

  In the above description, the generation of the prediction model (machine learning) and the prediction are performed by the management server 1, which is the same computer, but, for example, the management server 1 generates the prediction model and the generated prediction model is used by the user. After installing on the terminal 2, the user terminal 2 may perform prediction using the prediction model. That is, separate computers may execute the machine learning process and the process using the product of the machine learning.

  The above description is based on the premise that there is one industry for each user. For users who are engaged in business in multiple industries, a prediction model is generated and predicted for each industry. The predicted values for each industry are added together to make the predicted value for the entire user. Further, although the sales amount is predicted, values such as SG & A expenses (SG & A expenses and general administrative expenses) and variable cost ratios can also be predicted by the same method as described above.

  According to the present embodiment, it is possible to calculate a predicted value of sales applied to the environment of each store by artificial intelligence, and create a plan using the calculated integrated value of each store as a plan value.

Embodiment 4
The present embodiment relates to a form in which a predicted value to be used as a reference when determining a planned value of an item included in a business plan is obtained by using variation information. Fluctuation information is information that has an impact on business performance and is mainly obtained within the company. For example, the situation of the business partners collected by the salesperson and the salary situation based on the performance evaluation of the employees. The fluctuation information is collected before the plan is created and stored in the large capacity storage unit 14 or the like. Change information cannot be applied to higher-level items included in the management plan. Therefore, each item is drilled down to obtain a subordinate item to which variation information can be applied. The predicted value for each subordinate item is obtained using the variation information. The predicted value for each lower item obtained is drilled up to obtain the predicted value for each upper item. In the following description, if you drill down on sales and accumulate the sales for each supplier to create a forecast for sales, drill down on the cost of sales and accumulate the forecast for the costs for each purchased product. In the case of creating the forecast value of the cost of sales, the case of drilling down the personnel cost and accumulating the personnel cost of each employee to create the forecast value of the personnel cost is shown.

  FIG. 69 is an explanatory diagram showing an example of creating a sales forecast. To simplify the explanation, there are three customers. The 2019 column shows the sales for this term. The unit is, for example, one million yen. The performance trend column shows values representing sales trends acquired for each customer. Here, the percentage change in sales forecast from the current term to the next term is shown as a percentage. The next term column shows the forecast value of the next year's sales calculated from the sales of this term and the rate of change. Company A was 100 for the current term, but is expected to decrease by 5% in the next term, so the forecasted value for next term will be 95. Company B expects to increase by 10% in the next term from 200 in the current term, so the forecast value for the next term is 220. Similarly, Company C will have a 5% reduction from 300 this term, and the forecast for the next term will be 285. Therefore, the sales value of these customers is added up to give a predicted sales value of 600. Sales for this term are collected from SFA (Sales Force Automation), accounting system, billing system, POS system, etc. The business performance trend is obtained, for example, from the secular change of the data collected from these systems.

  FIG. 70 is an explanatory diagram showing an example of creating a sales cost forecast. In order to simplify the explanation, the cost of sales is only the cost for each purchased product, and the purchased products are three products. The 2019 column shows the cost for the current term. The unit is, for example, one million yen. The price trend column shows values representing price trends acquired for each purchased product. Here, the percentage change in the estimated cost from this term to the next term is shown as a percentage. The next term column shows the forecast value of the cost for the next term calculated from the cost for this term and the change rate. Product A had 10 in the current term, but is expected to decrease by 10% in the next term, so the forecast value for next term is 9. B product was 60 in this term, but is expected to increase by 5% in the next term, so the forecast value for next term is 63. Similarly, C product will increase by 5% from the current 40, and the forecast for the next fiscal year will be 42. Therefore, the estimated value of the cost of sales becomes 114 by adding up the costs for each of the purchased products. Costs are collected from the accounting system, ordering system, etc. The price trend is obtained, for example, from the secular change of the data collected from these systems.

  FIG. 71 is an explanatory diagram showing an example of personnel cost prediction. There are three employees to simplify the explanation. The 2019 column shows the total amount of salary for this term. The unit is, for example, one million yen. The increase / decrease rate column shows the expected increase / decrease rate of salary for each employee from this term to the next term in percentage. The next term column shows the predicted value of the total amount of payment for the next term calculated from the total amount of payment for this term and the rate of change. Employee A had a 10 in the current term, but is expected to decrease by 10% in the next term, so the forecasted value for the next term is 9. Employee B is expected to increase by 5% next year from 20 in the current term, so the forecast for next term is 21. Similarly, C employees will increase by 5% from the current 40, and the forecast for the next fiscal year will be 42. Therefore, the total amount of payment for each of these employees is added up, and the predicted value of the personnel cost becomes 72. The total amount paid is collected from attendance systems, payroll systems, accounting systems, etc. The rate of increase / decrease is obtained from, for example, the data collected from these systems, the personnel evaluation system, and the performance evaluation of each employee collected from the work evaluation system.

  FIG. 72 is an explanatory diagram showing an example of P / L based on predicted values. To simplify the description, only the predicted values shown in FIGS. 69 to 71 are created as the data source. The predicted value of sales is 600 shown in FIG. The cost of sales is 114 shown in FIG. Gross profit is 486 because it is the value obtained by subtracting the cost of sales (114) from the sales amount (600). The labor cost is 72 shown in FIG. Operating income is the gross profit (486) minus selling expenses and general control. Here, in order to simplify the explanation, the selling cost and the general administrative cost are only the value of the personnel cost. Therefore, the operating profit is 414 obtained by subtracting the personnel expenses (72) from the gross profit (486).

  The predicted value shown in FIG. 72 may be used as the planned value as it is. Alternatively, a value obtained by correcting the predicted numerical value may be used as the planned value. The correction may be performed by a person, or the management server 1 may refer to the statistical data or the like to make the correction.

  FIG. 73 is a flowchart showing a procedure example of the planned value calculation process. The planned value calculation process is a process corresponding to step S35 in FIG. The CPU 11 of the management server 1 acquires the variation information (step S401). The CPU 11 selects an item to be calculated (step S402). The CPU 11 calculates the predicted value (step S403). As shown in FIGS. 69 to 71, the predicted value is calculated from the actual value for the current term and the fluctuation information (business partner performance trend, purchase price trend, employee salary increase / decrease rate, etc.). The CPU 11 determines whether or not there is an unprocessed item (step S404). When the CPU 11 determines that there is an unprocessed item (YES in step S404), the process is returned to step S402, and a prediction value is calculated for the unprocessed item. When the CPU 11 determines that there are no unprocessed items (NO in step S404), it performs manual correction (step S405). The predicted value obtained by the CPU 11 is transmitted to the user terminal 2. The user confirms the predicted value displayed by the user terminal 2. The user makes the necessary corrections to the predicted values. The user terminal 2 sends the corrected value to the management server 1. The CPU 11 of the management server 1 sets the corrected value as the planned value. The CPU 11 outputs the planned value (step S406). For items that have not been manually corrected, the planned value is the predicted value.

  Creation of the following variation information will be described. FIG. 74 is an explanatory diagram showing an example of the prediction setting screen. The forecast setting screen shown in FIG. 74 is used when creating the business performance trend of a supplier, which is an example of change information. First, the performance trend for each customer is the predicted increase / decrease rate, which is the increase / decrease rate obtained from the sales for the past 6 periods. For example, the increase / decrease rate in 2019 based on 2014. When the user's 6th period transition is seen, if necessary, the value is corrected by directly editing the predicted increase / decrease rate column of the table. Further, the user can set the predicted increase / decrease rate by selecting the expected trend from the pull-down menu of the prediction column. For example, the 6th term increase / decrease rate of Company A is -3%, but the predicted column value is "slightly decreased", so the predicted increase / decrease rate is -5%. Since the forecast column for Company B is "flat", the forecast increase / decrease rate is the same value as the 6th quarter increase / decrease rate (+ 10%). Since the value in the forecast column for Company C is "decrease," the forecast increase / decrease rate is -5%, while the 6th term increase / decrease rate is 0%. The rule for obtaining the predicted increase / decrease rate from the 6th period increase / decrease rate and the value of the predicted column is set in advance. For example, if the predicted column is “slightly decreased”, the predicted increase / decrease rate is obtained by subtracting 2 points from the 6th period increase / decrease rate. For example, subtracting a predetermined value from the 6th period increase / decrease rate to obtain the predicted increase / decrease rate. Other calculation methods may be used. For example, the predicted increase / decrease rate may be a value obtained by multiplying the 6th period increase / decrease rate by a predetermined coefficient. FIG. 74 shows an example of a screen showing the business trend of the business partner, but the same setting screen is used for the price trend of the purchased product and the employee salary trend described above.

  FIG. 75 is a flowchart showing a procedure example of the prediction setting process. The prediction setting process is a process of setting the variation information used for calculating the predicted value. The CPU 11 of the management server 1 sends a list of items to the user terminal 2 (step S411). Examples of items are business performance trends, purchase price trends, and employee salary increase / decrease rates. The user selects the item to be set from the list of items. The CPU 21 of the user terminal 2 acquires the selection item (step S412). The CPU 21 transmits the selection item to the management server 1 (step S413). The CPU 11 of the management server 1 receives the selection item (step S414). The CPU 11 creates a list of selection items and sends it to the user terminal 2 (step S415). The CPU 21 of the user terminal 2 receives and displays the list (step S416). A display example of the list is the prediction setting screen of FIG. 74. The user makes changes as necessary. The CPU 21 acquires the changed content (step S417). The CPU 21 transmits the changed content to the management server 1 (step S418). The CPU 11 of the management server 1 receives the changed content (step S419). The change contents of the CPU 11 are reflected in the stored variation information (step S420).

  In the present embodiment, the predicted value that is the basis of the planned value is calculated using the variation information. As a result, it becomes possible to obtain a predicted value and a planned value that are more accurate than the planned value calculated from only past actual values.

Embodiment 5
The present embodiment relates to a form in which the prediction obtained in the fourth embodiment is corrected by external environment information. FIG. 76 is an explanatory diagram showing an example of the external environment information DB. The external environment information DB is stored in the large capacity storage unit 14 or the like. The external environment information DB includes an item string, a trend string, and a condition string. The item string stores items affected by the external environment. The trend column stores the trend of items. Here, the increase / decrease rate of the value of the item sequence is stored. The condition column stores the affected condition. The external environment information is created based on the information released by think tanks, statistical institutions, market research companies, information providers and the like.

  FIG. 77 is a flowchart showing another procedure example of the planned value calculation processing. Since the planned value calculation processing in the present embodiment is a modification of the planned value calculation processing in the fourth embodiment, FIG. 77 shows part of the processing. After calculating the predicted value (step S403), the CPU 11 of the management server 1 acquires the external environment information related to the item for which the predicted value is calculated (step S407). The CPU 11 corrects the predicted value based on the external environment information (step S408). For example, when the item is sales and the type of business of the user is XX, the CPU 11 corrects the predicted value to a value 5% lower by the external environment information shown in FIG. The CPU 11 executes step S404 and subsequent steps in FIG. 73.

  Next, the creation of external environment information will be described. FIG. 78 is a flowchart showing an example of the procedure of the external environment information creation processing. The CPU 11 of the management server 1 acquires the user ID of the user who uses the external environment information (step S431). The CPU 11 acquires trend information released by think tanks and the like, economic indicators released by statistical institutions (hereinafter referred to as external information) (step S432). The CPU 11 filters by industry (step S433). The CPU 11 searches the user DB 141 using the user ID as a search key, and acquires the user's business category. It is determined whether the user's industry is affected by the external information. It is assumed that the rule for judging is stored in advance. If the user's business type is not affected, the CPU 11 ends the process. If the user's industry is affected, the CPU 11 continues the process. The CPU 11 performs area filtering (step S434). It is determined whether or not the geographical range affected by the external information and the business range of the user overlap. If the business range of the user is not included in the affected geographical range, the CPU 11 ends the process. If the business range of the user is included in the affected geographical range, the CPU 11 continues the process. The CPU 11 determines the timing (step S435). The CPU 11 determines whether to be immediately affected by the external information or to be delayed. It is assumed that the rule for judging is stored in advance. When the CPU 11 determines that the influence indicated by the external information is affected during the planning period (for example, during the next period), the affected period is stored in the temporary storage area provided in the RAM 13 or the large-capacity storage unit 14, Continue processing. When the CPU 11 determines that it is not affected by the external information during the planning period, the CPU 11 ends the process. The CPU 11 makes an impact determination (step S436). The CPU 11 determines the degree of being affected by the external information. It is assumed that the rule for judging is stored in advance. The degree of influence is represented by a weighting coefficient, for example. The CPU 11 stores the degree of influence in the temporary storage area. Based on the above processing, the CPU 11 generates external environment information and stores it in the mass storage unit 14 (step S437). The CPU 11 ends the process. In the present embodiment, it is assumed that the user DB 141 stores the business type of the user and the business range of the user in association with the user ID. The external environment information creation process is performed for each user and for each item including the business plan.

  Next, an example of creating the external environment information will be shown. FIG. 79 is an explanatory diagram showing an example of generating external environment information from external information. FIG. 79 shows a procedure for generating external environment information from the external information “expected increase in freight cost due to soaring crude oil price”. In the business type filtering, for example, the determination is made according to the determination rule 791. If the user's business type is the manufacturing industry, the processing is continued because it directly affects the business. If the user's business type is the service business, it is not affected and the process ends. In the area filtering, the determination is made according to the determination rule 792, for example. Since shipping costs are affected nationwide, all users who have reached area filtering are eligible. In the timing determination, for example, the timing affected by the determination rule 793 is determined. Since the shipping fee has a time lag of 1 to 2 months, for example, when the fiscal year is from April to March of the following year, the affected month is 10 months from June to March of the following year. In the impact determination, for example, the degree of influence is determined by the determination rule 794. For example, the weight of the purchase cost indirectly affected by the increase in the shipping cost is 0.5. The external environment information 795 is generated from the above processing and stored. The freight cost will increase by 5%, and the indirectly affected purchase cost will increase by 2.5% due to the weighting factor of 0.5.

  Next, an example of correcting the predicted value will be shown. FIG. 80 is an explanatory diagram showing an example of correction based on the external environment information. The table 811 is an example of P / L. It is similar to FIG. Table 812 shows the corrected values. Table 813 shows an example of external environment information. There is information that sales are down 5% as external environmental information. In the correction of the sales amount, the predicted value is multiplied by 0.95 (= 1-0.05). The corrected sales value is 600 × 0.95, which is 570. Since there is external environmental information that the purchase cost is increased by 2.5%, the estimated value is multiplied by 1.025 (= 1 + 0.025) in the correction of the cost of sales. The corrected cost of sales is 114 × 1.025, which is 116. Gross profit is 570-116 = 454 after correction. Since there is external environmental information that the labor cost will increase by 10%, the compensation value of the labor cost is 1.1 (= 1 + 0.1). The corrected personnel cost is 72 x 1.1, which is 79. When amending, round off below the decimal point. The corrected operating profit is 454-79 = 375. Of the external environment information shown in the table 813, the sales amount can be created based on, for example, a market trend survey conducted by think tanks. Trends in personnel costs can be created, for example, from the basic statistical survey on wage structure conducted by the government. An example of creating a trend of purchase cost is shown in FIG. 79.

  In the present embodiment, the predicted value is corrected by the external environment information, so that a more realistic plan can be created. It is expected that accuracy will also improve.

  Since the external environment information can be created from various external information, different external information may be obtained for the same item. In such a case, the following processing can be considered. For example, when you get inconsistent external information such as A think tank: personnel expenses increase by 10% due to lack of manpower, B think tank: personnel expenses decrease by 1% due to deterioration of business performance due to recession. Discard. Information 1: The wholesale prices are expected to increase by 10% due to rising labor costs, and Information 2: Wholesale prices are expected to increase by 5% due to rising crude oil prices. , For example, a total increase of 15%. A think tank: Personnel expenses increased by 10% due to labor shortage, C think tank: Personnel expenses increased by 5% due to labor shortage, etc. As a result, personnel expenses will increase by 7.5%. The above is an example, and other processing may be performed.

The technical features (constituent elements) described in each embodiment can be combined with each other, and by combining them, new technical features can be formed.
The embodiments disclosed this time are to be considered as illustrative in all points and not restrictive. The scope of the present invention is shown not by the meanings described above but by the claims, and is intended to include meanings equivalent to the claims and all modifications within the scope.

100 plan management system 1 management server 11 CPU
12 ROM
13 RAM
14 Mass Storage 141 User DB
142 Accounting DB
143 reservation DB
145 attendance DB
146 Ordering DB
147 Logic DB
148 Actual DB
149 Report DB
14A Alert DB
15 communication unit 16 reading unit 1P control program 1a portable storage medium 1b semiconductor memory 2 user terminal 21 CPU
22 ROM
23 RAM
24 communication unit 25 input unit 26 display unit 27 large-capacity storage unit 2P control program 3 administrator terminal 4 other system

Claims (6)

Acquire the value of each of the multiple description items for multiple main items including sales and cost of sales related to management,
To the learned model that outputs the predicted value of the plurality of main items when the value of each of the plurality of description items is input, input the value of each of the obtained plurality of description items,
Obtaining predicted values of the plurality of main items output by the learned model ,
Select the expected change of the main items, output questions to answer,
Acquiring the selection information of the likelihood of the change for the question, and associated with the acquired selection information, obtain a calculation logic including a coefficient that has converted the likelihood of the change,
Correct the predicted values of the main items based on the acquired calculation logic,
Based on the corrected predicted values of the main items, calculate an index value including at least one of the sales or gross profit constituting the business plan,
A plan creation program characterized by causing a computer to perform processing to output the created index values in a graph displayable manner. Acquiring actual data for each of a plurality of explanatory items for a plurality of main items including sales and cost of sales related to management,
The plan creation program according to claim 1, wherein actual data of each of the plurality of description items is input to the learned model .   Acquire internal environment information that affects the above-mentioned items,
  Correct the predicted value of the main item corresponding to the explanation item based on the acquired internal environment information.
  The plan creation program according to claim 1 or 2, characterized in that. Acquire external environment information that affects the main items ,
The plan creation program according to any one of claims 1 to 3 , wherein the predicted value of the main item is corrected based on the acquired external environment information. A first obtaining unit configured to obtain the value of each of the plurality of description items for a plurality of main items, including sales and cost of sales related to management,
An input unit that inputs the value of each of the acquired plurality of description items to the learned model that outputs the predicted value of the plurality of main items when the value of each of the plurality of description items is input.
A second acquisition unit that acquires predicted values of the plurality of main items output by the learned model ;
A question output unit that outputs a question to be answered by selecting the likelihood of change in the main item,
A third acquisition unit that acquires selection information of an answer to the question and that is associated with the acquired selection information and that acquires a calculation logic that includes a coefficient that converts the probability of change.
A correction unit that corrects the predicted value of the main item based on the acquired calculation logic,
A calculation unit that calculates an index value that includes at least one of sales or gross profit making up the management plan, based on the predicted values of the corrected main items,
An output unit for outputting the created index value so that it can be displayed as a graph . Computer
Acquire the value of each of the multiple description items for multiple main items including sales and cost of sales related to management,
To the learned model that outputs the predicted value of the plurality of main items when the value of each of the plurality of description items is input, input the value of each of the obtained plurality of description items,
Obtaining predicted values of the plurality of main items output by the learned model ,
Select the expected change of the main items, output questions to answer,
The selection information of the answer to the question is acquired, and the calculation logic including the coefficient that is associated with the acquired selection information and converted the likelihood of the change is acquired,
Correct the predicted values of the main items based on the acquired calculation logic,
Based on the corrected predicted values of the main items, calculate an index value including at least one of the sales or gross profit constituting the business plan,
A plan creation method characterized by outputting the created index values in a graph displayable manner.

Images