Business support systems, business support methods, and programs

Abstract

The business system is evaluated accurately at the business requirements stage. [Solution] The business support system includes a machine learning unit that performs machine learning based on a dataset including business requirements that include multiple tasks and the work hours or work period of the tasks to generate a trained model, and a prediction processing unit that, when business requirements that are expected to change are input, predicts at least the work hours or work period for each task included in the business requirements based on the trained model.

Description

Technical Field

[0005] , , ,

[0004] , , , , , , , , , , ,

[0006] ,

[0001] The present disclosure relates to a business support system, a business support method, and a program.

Background Art

[0002] There is known a support system that outputs work time and the like using machine learning to support business operations (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, in recent years, due to the amendment of the Electricity Business Act, the former general electricity business operators have been legally separated into retail electricity business operators, general transmission and distribution business operators, and power generation business operators. Along with this, a large number of operators and systems are involved, and the construction of systems for power services or related operations is required. However, for example, in the business system of a transmission and distribution business operator, when making changes or improvements to the system, it is difficult to accurately estimate the work hours or work time from the business requirements, and it is difficult to accurately evaluate the business system at the stage of business requirements.

[0005] The present disclosure has been made to solve the above problems, and an object thereof is to provide a business support system, a business support method, and a program capable of accurately evaluating a business system at the stage of business requirements.

Means for Solving the Problems

[0006] To solve the above problem, one aspect of this disclosure is a business support system comprising: a machine learning unit that generates a trained model by performing machine learning on a dataset including business requirements that include multiple tasks and the work hours or work period of the tasks; and a prediction processing unit that, when business requirements that are expected to change are input, predicts at least the work hours or work period for each task included in the business requirements based on the trained model.

[0007] Furthermore, one aspect of this disclosure is a business support method in which a machine learning unit performs machine learning on a dataset including business requirements that include multiple tasks and the work hours or work period of the tasks to generate a trained model, and a prediction processing unit, when business requirements that are expected to change are input, predicts at least the work hours or work period for each task included in the business requirements based on the trained model.

[0008] Furthermore, one aspect of this disclosure is a program that causes a computer to perform a machine learning step to generate a trained model by performing machine learning on a dataset including business requirements that include multiple tasks and the work effort or work period of the tasks, and when business requirements that are expected to change are input, a prediction step to predict at least the work effort or work period for each task included in the business requirements based on the trained model. [Effects of the Invention]

[0009] According to this disclosure, business systems can be accurately evaluated at the business requirements stage. [Brief explanation of the drawing]

[0010] [Figure 1] This is a functional block diagram showing an example of a business support system according to this embodiment. [Figure 2] This figure shows an example of the processing flow for each operator in the electricity business. [Figure 3] This figure shows an example of the business flow in this embodiment. [Figure 4] This figure shows an example of machine learning processing for contract registration / modification in this embodiment. [Figure 5] This figure shows an example of machine learning processing for power consumption processing in this embodiment. [Figure 6] This figure shows an example of machine learning processing for various fee calculation processes in this embodiment. [Figure 7] This figure shows an example of machine learning processing for invoice processing for businesses in this embodiment. [Figure 8] This figure shows an example of the prediction process in this embodiment. [Figure 9] This figure shows an example of the display of the comparison process in the business flow in this embodiment. [Figure 10] This is the first figure showing an example of a comparative display of the prediction processing results in this embodiment. [Figure 11] This is the second figure, showing an example of a comparative display of the prediction processing results in this embodiment. [Figure 12] This flowchart shows an example of machine learning processing in a business support system according to this embodiment. [Figure 13] This flowchart shows an example of the evaluation process for the business flow of the business support system according to this embodiment. [Figure 14] This figure shows an example of how the prediction processing results are displayed in the first modified example of this embodiment. [Figure 15] This figure shows an example of how the prediction processing results are displayed in a second modified example of this embodiment. [Figure 16] This figure shows an example of how the prediction processing results are displayed in a third modified example of this embodiment. [Figure 17] This figure shows an example of how the prediction processing results are displayed in a fourth modified example of this embodiment. [Figure 18] This figure shows an example of how the prediction processing results are displayed in a fifth modified example of this embodiment. [Figure 19] This figure shows an example of how the prediction processing results are displayed in a sixth modified example of this embodiment. [Figure 20]It is a block diagram showing an example of the hardware configuration of the business support server according to this embodiment.

Embodiment for Carrying Out the Invention

[0011] Hereinafter, a business support system and a business support method according to an embodiment of the present disclosure will be described with reference to the drawings.

[0012] FIG. 1 is a schematic block diagram showing an example of a business support system 100 according to this embodiment. As shown in FIG. 1, the business support system 100 includes a business support server 10, a user terminal 14, a power transmission and distribution business system 15, a power meter 20, a meter reading terminal 21, a power generation business system 30, and a power retail business system 40.

[0013] The business support system 100 is a system that provides business support for the power transmission and distribution business among power-related businesses, which are operators related to the electricity business. Note that the power-related businesses include a power transmission and distribution operator 1, a power generation operator 3, and a power retail operator 4. In addition to the power-related businesses, there are power consumers 2 who receive and consume electric power.

[0014] The power consumer 2 is, for example, a consumer who uses electricity as a power retail operator 4, including not only ordinary households but also consumers in enterprises. The power consumer 2 has a power meter 20.

[0015] The power meter 20 is, for example, a smart meter that measures the amount of electric power used by the power consumer 2 and transmits the measured power amount data to the power transmission and distribution business system 15 of the power transmission and distribution operator 1 via the network NW1.

[0016] The meter reading terminal 21 is a terminal device used by a meter reading worker who reads the amount of electric power used, and transmits the read power amount data to the power transmission and distribution business system 15 of the power transmission and distribution operator 1 via the network NW1.

[0017] Power generation company 3 is a business that generates electricity and supplies electricity (power). Power generation company 3 has a power generation business system 30. The power generation system 30 controls the generator and, according to the contract, performs business processing for the power generation business that supplies electricity.

[0018] Electricity retailer 4 is a retail electricity provider that supplies electricity to customers (electricity consumers 2) with whom it has a contract for electricity supply, while ensuring the necessary supply capacity. Electricity retailer 4 has an electricity retail business system 40. The electricity retail business system 40 executes the business processes of the electricity retail business operator 4.

[0019] The power transmission and distribution operator 1 is a business that supplies electricity received from the power generation operator 3 to the power retail operator 4, and as a grid operator, it performs final supply and demand adjustments and construction and maintenance of the power transmission and distribution network. The power transmission and distribution operator 1 has a business support server 10, a user terminal 14, and a power transmission and distribution business system 15.

[0020] The power transmission and distribution system 15 executes the business processes of the power transmission and distribution operator 1. Details of the business processes of the power transmission and distribution operator 1 will be described later with reference to Figure 2. The user terminal 14 is a terminal device owned by the power transmission and distribution company 1, and is, for example, a personal computer (PC) or a tablet terminal. The user terminal 14 can connect to the business support server 10 via the network NW1 and is used to operate the business support server 10.

[0021] Now, referring to Figure 2, we will explain the relationships and business processes among the various operators in the power business. Figure 2 shows an example of the processing flow for each business operator in the electricity business.

[0022] As shown in Figure 2, first, electricity consumer 2 applies for / changes electricity supply services (step S101) and sends an application / change notification to electricity retailer 4 (step S102).

[0023] Next, the electricity retailer 4 executes the service contract and installation process with the electricity consumer 2 (step S103), and notifies the transmission and distribution company 1 of the connection supply contract (step S104).

[0024] On the other hand, power generator 3 notifies transmission and distribution operator 1 of the power generation adjustment supply contract with transmission and distribution operator 1 (step S105).

[0025] Next, the power transmission and distribution operator 1 performs a contract registration / modification process (step S106) and requests the electrical contractor to perform the work (step S106). The contract registration / modification process is a business process that takes application details (customer information, business information, contract type, etc.) as input information and outputs the application result and various internal requests (completion requests, etc.) based on said input information.

[0026] Next, the electrical contractor performs the electrical work in response to the request from the power transmission and distribution company 1 (step S108), and notifies the power consumer 2 of the completion of the work (step S108). As a result, electricity consumer 2 begins using electricity (step S109).

[0027] Next, the power transmission and distribution operator 1 collects electricity usage data from the electricity consumer 2 (step S110), and performs electricity processing based on the collected electricity usage data (step S111). Electricity processing is a business process that takes meter values ​​(monthly usage, 30-minute values, 15-minute values, 5-minute values) and planned values ​​as input information, and outputs information necessary for various calculations such as electricity classification, maximum DM, and contracted power based on said input information.

[0028] Next, the power transmission and distribution operator 1 performs various charge calculation processes (processing in step S112) and invoice processing for businesses (processing in step S113), and in steps S114 and S115, notifies the power retail operator 4 and the power generation operator 3 of the invoices.

[0029] Various fee calculation processes are business processes that take customer information, contract information, and calculation parameters as input information, and output fee calculation results and reports based on that input information. Furthermore, invoice processing for businesses involves taking documents and payment requests as input information and outputting responses such as whether or not the documents and payment requests have been received based on that input information.

[0030] Next, in steps S116 to S118, the electricity retailer 4 performs the calculation of charges and the processing of customer billing, and notifies the electricity customer 2 of the invoice.

[0031] Next, in steps S119 to S121, the electricity retail business operator 4 executes the business payment request process and notifies the transmission and distribution business operator 1 and the power generation business operator 3. Furthermore, power generation company 3 executes the business payment request process (step S122) and notifies transmission and distribution company 1 (step S123).

[0032] In the processing flow shown in Figure 2, steps S103, S104, and steps S116 to S121 performed by the electricity retail business operator 4 are primarily executed using the electricity retail business system 40.

[0033] Furthermore, the processes performed by the power generation operator 3 in steps S105, S113, and S114 are mainly executed using the power generation business system 30. Furthermore, the processes performed by the power transmission and distribution operator 1 from step S106, step S107, and steps S110 to S115 are mainly executed using the power transmission and distribution system 15.

[0034] Furthermore, in this embodiment, the business processes GS supported by the business support system 100 (business support server 10) are contract registration / modification processing by the power transmission and distribution company 1 (processing in step S106), power consumption processing (processing in step S111), various charge calculation processing (processing in step S112), and invoice processing for businesses (processing in step S113).

[0035] Contract registration / modification processing, electricity usage processing, various charge calculation processing, and business invoice processing each correspond to business processes, and each process is defined by business requirements and processed based on those requirements. Furthermore, business process GS, which includes contract registration / modification processing, electricity usage processing, various charge calculation processing, and business invoice processing, shall be processed based on the overall business requirements, which include each of the individual business requirements. In other words, the business requirements for business process GS are business requirements that include multiple processes related to electricity charge calculation.

[0036] In this embodiment, the business requirements specifically describe, for example, the objectives that the service or business should achieve, and the functions and conditions necessary to achieve them. This includes business flows that show the specific processing of each business process or the overall business process (business process GS). In this embodiment, a business flow will be used to explain an example of business requirements.

[0037] Returning to the explanation of Figure 1, the business support server 10 is a server device that supports business by outputting, for example, the work hours or work period of the business (business processing) described above, and can communicate with the user terminal 14, the power transmission and distribution business system 15, the power generation business system 30, and the power retail business system 40 via the network NW1. The business support server 10 comprises a network communication unit 11, a server storage unit 12, and a server control unit 13.

[0038] The NW communication unit 11 is a functional unit implemented by a communication device such as a network adapter. The NW communication unit 11 connects to the network NW1 and communicates with, for example, a user terminal 14, a power transmission and distribution system 15, a power generation system 30, and a power retail system 40.

[0039] The server storage unit 12 stores various information used by the business support server 10. The server storage unit 12 includes a learning data storage unit 121, a model storage unit 122, a linked data storage unit 123, a past performance storage unit 124, a business flow storage unit 125, an evaluation condition storage unit 126, a proposed content storage unit 127, and an analysis result storage unit 128.

[0040] The training data storage unit 121 stores the dataset, which is the training data used for the machine learning processing described later. Details of the training data dataset will be described later.

[0041] The model storage unit 122 stores trained models generated by machine learning processing, which will be described later. The model storage unit 122 stores trained models corresponding to each of the following business processes: contract registration / modification processing, electricity usage processing, various fee calculation processing, and invoice processing for businesses.

[0042] The linked data storage unit 123 stores, for example, linked data between a power transmission and distribution operator 1 (power transmission and distribution system 15), a power generation operator 3 (power generation system 30), and a power retail operator 4 (power retail system 40). The linked data includes information about contracts with power consumers 2, power generation operators 3, and power retail operators 4 (for example, application type information). The linked data storage unit 123 stores, as linked data, customer information, which is information about the contracted power consumer 2; operator information, which is information about power generation operators 3 and power retail operators 4; and contract information, such as contract type.

[0043] Customer information includes, for example, the number of contracts (processing volume) for electricity consumer 2, while business information includes, for example, the number of contracts (processing volume) for power generation company 3 and electricity retail company 4. Contract information also includes meter reading date information, rate adjustment schedule (rates by date, imbalance charges, etc.), contracted electricity volume, etc.

[0044] The past performance storage unit 124 stores past performance information for each business (each business process) of the business process GS. Past performance information includes, for example, worker information, work costs, past costs, and information on past irregular handling. Worker information also includes, for example, the worker's years of experience, work proficiency, and skill information, while work costs include, for example, the worker's cost and the cost of materials necessary for the work.

[0045] Furthermore, past costs include, for example, the number of workers and man-hours worked in the past. Past irregular response includes the nature of the irregularity (irregular event), the number of workers and man-hours required to respond to the irregularity. Irregular events include, for example, power outages, deficiencies in power data, deficiencies in the coordination of power data between power-related businesses, and deficiencies in power-related contract data.

[0046] Furthermore, past performance information may include historical information that associates work identification information for each worker with work hours or work period and calculation period information. Furthermore, past performance information may include, for example, the skills required for the job and the work hours for each skill.

[0047] The business flow storage unit 125 stores the business flow of the business process GS described above. The business flow storage unit 125 stores, for example, the business flow for each business of the business process GS. Here, the business flow is a flow that shows the processing of a specific business, such as the one shown in Figure 3.

[0048] Figure 3 shows an example of the business flow in this embodiment. The business flow FL1 shown in Figure 3 is a flow that illustrates the processing of reception tasks. The business process flow storage unit 125 stores business processes for each business process, as shown in Figure 3. The business processes stored by the business process flow storage unit 125 may include, for example, business processes that are expected to change due to system changes, business processes that have been or are currently in use, or hypothetical business processes that have been considered experimentally.

[0049] The evaluation condition storage unit 126 uses a trained model, described later, to predict the workload or duration of a task and stores the evaluation conditions for evaluating the task. The evaluation conditions include, for example, the business flow to be evaluated (including new business flows), costs (target cost, workload, etc.), processing time (target duration), number of workers, and worker information. The evaluation conditions may also include meter reading date information, rate adjustment schedule, transmission and distribution region or group, and information indicating irregular events.

[0050] The proposal content storage unit 127 stores the proposal content, which is the evaluation result (prediction result) of the business process. The proposal content includes, for example, the estimated cost, the estimated work time (an example of the work period), the appropriate number of workers (an example of the work man-hours), and the appropriate skill level.

[0051] The analysis result storage unit 128 stores the analysis results obtained when the analysis processing unit 135, which will be described later, analyzes the prediction results. The analysis results include, for example, various comparison results based on the prediction results of the prediction processing unit 134 (for example, work hours or work period), which will be described later, and output information (display information) generated by the output control unit 136, which will be described later.

[0052] The server control unit 13 is a functional unit that is realized by, for example, causing a processor including a CPU (Central Processing Unit) to execute a program stored in the server storage unit 12. The server control unit 13 executes various processes of the business support server 10. The server control unit 13 comprises a data collection unit 131, a training data generation unit 132, a machine learning unit 133, a prediction processing unit 134, an analysis processing unit 135, and an output control unit 136.

[0053] The data collection unit 131 (an example of a data collection unit) collects various types of information for generating training data from the power transmission and distribution business system 15, the power generation business system 30, and the electricity retail business system 40. The data collection unit 131 collects, for example, past performance information, linked data, business flow, meter values / planned values, charge calculation information (charge menu), and invoice processing information for businesses.

[0054] The meter values / planned values ​​include, for example, interval data (e.g., 5-minute, 10-minute, and 30-minute values), monthly electricity usage, and planned values ​​(daily data linkage count, number of missing data points, etc.). The billing calculation information includes, for example, customer information (contractor information), contract information / schedule information, and calculation parameters (e.g., input information for the number of items subject to billing calculation per billing menu per day / schedule).

[0055] Furthermore, the invoice processing information for businesses includes, for example, document information and payment request forms (such as input information on the number of documents / payment request forms per date / day). The data collection unit 131 may, for example, collect for each worker work identification information and historical information (an example of past performance information) relating to work hours or work period.

[0056] Furthermore, the data collection unit 131 stores the collected past performance information in the past performance storage unit 124. The data collection unit 131 also stores the collected linked data in the linked data storage unit 123. Additionally, the data collection unit 131 stores the collected business flow in the business flow storage unit 125. The data collection unit 131 may also directly store some of the collected information in the learning data storage unit 121.

[0057] The learning data generation unit 132 generates a learning data set based on the data collected by the data collection unit 131. For example, for each worker, the learning data generation unit 132 generates a data set by aggregating the work hours or work period for each task indicated by the work identification information, based on the work identification information that identifies the task and historical information of work hours or work period.

[0058] A dataset may include, for example, a business flow (business requirements) that includes multiple tasks, and the work hours or duration of the tasks. Furthermore, the dataset may also include the meter reading date for electricity consumption, and further include the rate adjustment schedule. In other words, a dataset may include, for example, a business flow, the work hours or duration of the tasks, and calculation period information (such as the meter reading date for electricity consumption and the rate adjustment schedule), which is time information necessary for calculating electricity charges.

[0059] Furthermore, the dataset may also include, for example, the skills required for the job and the effort required for each skill. Furthermore, the dataset may include, for example, a workflow corresponding to the invoice creation process for transmission charges (e.g., processing invoices for businesses) and the work hours or duration of the process, or it may include, for example, a workflow for electricity charge calculation (e.g., various charge calculation processes) and the work hours or duration of the process.

[0060] Furthermore, the dataset may include, for example, information indicating an irregular event and the workload or duration of the work performed when the irregular event occurs. The training data generation unit 132 stores the generated dataset, which is the training data, in the training data storage unit 121.

[0061] The machine learning unit 133 performs machine learning based on the training data generated by the training data generation unit 132 to generate a trained model. In other words, the machine learning unit 133 performs machine learning based on the training data stored in the training data storage unit 121 to generate a trained model for each business process.

[0062] The machine learning unit 133 generates a trained model by performing machine learning based on a dataset that includes, for example, a business flow (business requirements) containing multiple tasks and the work hours or duration of the tasks. Furthermore, the machine learning unit 133 generates a trained model by performing machine learning based on a dataset that includes a business flow, the work hours or duration of the tasks, and calculation period information, which is time information necessary for calculating electricity charges.

[0063] Here, with reference to Figures 4 to 7, we will explain specific examples of machine learning processing corresponding to each business process performed by the machine learning unit 133. Figure 4 shows an example of machine learning processing for contract registration / modification in this embodiment.

[0064] As shown in Figure 4, the machine learning unit 133 performs machine learning processing on the contract registration / modification process based on training data from a dataset that includes past performance (worker information, work costs, past man-hours, and past handling of irregularities), application type (linked data), and business flow, and generates a trained model for the contract registration / modification process. Here, the application type (linked data) is, for example, contract information such as customer information, business operator information, and contract type.

[0065] The machine learning unit 133 performs machine learning in such a way that it minimizes an evaluation function consisting of a cost model and a processing time model, for example, as shown in equation (1) below.

[0066] Evaluation function = Q × work cost model + R × processing time model … (1)

[0067] Here, Q and R are weight parameters. The machine learning unit 133 stores the generated trained model for contract registration / modification processing in the model storage unit 122.

[0068] Figure 5 shows an example of machine learning processing for power consumption processing in this embodiment.

[0069] As shown in Figure 5, the machine learning unit 133 performs machine learning processing on the business process of power consumption processing based on training data from a dataset that includes past performance (worker information, work costs, past man-hours, and past handling of irregularities), meter values / planned values, and business flow, and generates a trained model for power consumption processing. Here, meter values / planned values ​​include, for example, interval data (e.g., 5-minute values, 10-minute values, 30-minute values), monthly power consumption, and planned values ​​(daily number of data linkages, number of missing data, etc.).

[0070] The machine learning unit 133 performs machine learning in such a way that it minimizes an evaluation function consisting of a cost model and a processing time model, as shown in equation (1) above. The machine learning unit 133 stores the generated trained model for power consumption processing in the model storage unit 122.

[0071] Figure 6 shows an example of machine learning processing for various fee calculation processes in this embodiment.

[0072] As shown in Figure 6, the machine learning unit 133 performs machine learning processing on various fee calculation processes based on training data from a dataset that includes past performance (worker information, work costs, past man-hours, and past handling of irregularities), fee calculation information, and business flow, and generates trained models for various fee calculation processes. Here, fee calculation information includes, for example, customer information (contractor information), contract information / schedule information, and calculation parameters (for example, input information on the number of items to be calculated for each fee menu per schedule / day).

[0073] The machine learning unit 133 performs machine learning in such a way that it minimizes an evaluation function consisting of a cost model and a processing time model, as shown in equation (1) above. The machine learning unit 133 stores the trained models generated for various fee calculation processes in the model storage unit 122.

[0074] Figure 7 shows an example of machine learning processing for invoice processing for businesses in this embodiment.

[0075] As shown in Figure 6, the machine learning unit 133 performs machine learning processing on business invoice processing based on training data from a dataset that includes past performance (worker information, work costs, past man-hours, and past handling of irregularities), business invoice processing information, and business flow, and generates a trained model for business invoice processing. Here, business invoice processing information includes, for example, document information and payment request forms (input information such as the number of documents / payment request forms per date / day).

[0076] The machine learning unit 133 performs machine learning in such a way that it minimizes an evaluation function consisting of a cost model and a processing time model, as shown in equation (1) above. The machine learning unit 133 stores the trained model for processing invoices for businesses that it has generated in the model storage unit 122.

[0077] Returning to the explanation of Figure 1, the prediction processing unit 134 predicts the amount of work or the duration of work for each task based on the trained model generated by the machine learning unit 133. The prediction processing unit 134 predicts the amount of work or the duration of work for a task based on evaluation conditions input via, for example, the user terminal 14 and the trained model stored in the model storage unit 122.

[0078] For example, when a business flow that is expected to change is input, the prediction processing unit 134 predicts, based on a trained model, at least the amount of work effort or work duration for each task included in the business flow (for example, the business flow corresponding to the overall business process GS).

[0079] Furthermore, the prediction processing unit 134, for example, when the workflow of the task to be predicted is input, predicts the work effort or work period of the task to be predicted based on the trained model. The prediction processing unit 134 may also predict the daily work effort of the task to be predicted.

[0080] Furthermore, the prediction processing unit 134, for example, when a business flow corresponding to the business to be predicted and the meter reading date are input, predicts the work effort or work period of the business to be predicted based on the trained model.

[0081] Furthermore, the prediction processing unit 134, for example, when a business flow and a fee adjustment schedule corresponding to the business to be predicted are input, predicts the work effort or work period of the business to be predicted based on the trained model.

[0082] Furthermore, the prediction processing unit 134, for example, when a business flow corresponding to the business to be predicted is input, predicts the work effort or work period for each skill required for the business to be predicted, based on the trained model.

[0083] Furthermore, the prediction processing unit 134, for example, when a region or group and a business flow corresponding to the predicted business are input, predicts the work effort or work period for the predicted business in that region or group based on the trained model.

[0084] Furthermore, if information indicating an irregular event is input, the prediction processing unit 134 predicts, for example, the amount of work required or the duration of work if such an irregular event occurs, based on the trained model.

[0085] Furthermore, if power outage information (such as whether or not there is a power outage, the duration of the power outage, etc.) is input, the prediction processing unit 134 predicts the workload or duration of work in the event of a power outage based on the trained model. Also, if information about deficiencies in power data (such as the number of delays, the number of out-of-order items, the number of items that cannot be linked, etc.) is input, the prediction processing unit 134 predicts the workload or duration of work in the event of deficiencies in power data based on the trained model.

[0086] Furthermore, if information such as a data linkage failure between power-related businesses is input, the prediction processing unit 134 predicts the workload or duration of work that would occur if a data linkage failure between power-related businesses occurs, based on the trained model. Also, if information such as a contract data deficiency is input, the prediction processing unit 134 predicts the workload or duration of work that would occur if a contract data deficiency occurs, based on the trained model.

[0087] Now, with reference to Figure 8, an example of the prediction process in this embodiment will be described. Figure 8 shows an example of the prediction process in this embodiment.

[0088] As shown in Figure 8, the prediction processing unit 134 takes evaluation conditions and business flow as input data and predicts, for example, estimated costs, appropriate number of workers, estimated work time, appropriate skill level, etc., based on trained models (trained model for contract registration / modification processing, trained model for power consumption processing, trained model for various charge calculation processing, and trained model for invoice processing for businesses).

[0089] Here, the evaluation criteria include, for example, cost, processing time, number of workers, and worker information. Some of the evaluation criteria, such as cost, processing time, number of workers, and worker information, may be entered. In this case, pre-set standard values ​​are used for any information that is not entered. The prediction processing unit 134 stores the prediction results in the proposed content storage unit 127, for example.

[0090] Returning to the explanation of Figure 1, the analysis processing unit 135 performs various analysis processes based on the prediction results of the prediction processing unit 134. The analysis processing unit 135 changes the setting conditions, including the business flow, and compares them with the work hours or work period of the business predicted by the prediction processing unit 134, and outputs the results to the output control unit 136, which will be described later.

[0091] The analysis processing unit 135, for example, compares the work hours or duration of the business process for the business process before the change with the work hours or duration of the business process for the business process after the change and outputs the result to the output control unit 136. The analysis processing unit 135, for example, compares the work hours or duration of the business process for the business process after the change with the work hours or duration of the business process for a similar business process that is similar to the business process after the change and outputs the result to the output control unit 136.

[0092] The analysis processing unit 135 stores the analysis results, such as comparison results, in the analysis result storage unit 128. The analysis processing unit 135 may also, for example, compare the work hours or work period for multiple modified business flows and output the results to the output control unit 136.

[0093] Figures 9 to 11 show examples of the display of the comparison processing results by the analysis processing unit 135. Figure 9 shows an example of the display of the comparison process in the business flow in this embodiment. The example shown in Figure 9 illustrates display screen G1, which compares a similar business flow FL2 with an existing business flow / new business flow FL3. Display screen G1 is shown, for example, on the display unit (not shown) of a user terminal 14 connected to the business support server 10 via network NW1.

[0094] On display screen G1, a similar business flow FL2 is displayed for comparison with the existing business flow / new business flow FL3, and the differences in the business flows are highlighted with hatching or other methods. The analysis processing unit 135 outputs the comparison results, as shown on display screen G1, to the output control unit 136.

[0095] Figures 10 and 11 show examples of comparative displays of prediction processing results in this embodiment. The display screen G2 shown in Figure 10 is an example of a display that compares and displays two items: similar business flow content and proposed content (for example, new business flow content).

[0096] Furthermore, the display screen G3 shown in Figure 11 shows an example of a display that compares three items: similar business flow content, past performance (for example, current business flow content), and proposed content (for example, new business flow content).

[0097] The analysis processing unit 135 outputs comparison results, such as those shown on display screens G2 and G3 in Figures 10 and 11, to the output control unit 136. In Figures 10 and 11, the content displayed in display sections OP1 and OP2 is shown, for example, by specifying options in the setting conditions.

[0098] Furthermore, the numerical values ​​shown on display screens G2 and G3 may be displayed using radio buttons on the screen for "overall average" or "daily average," and the output control unit 136 may control the numerical values ​​of the displayed items (overall average or daily average) according to the user's selection.

[0099] Returning to the explanation of Figure 1, the output control unit 136 outputs output information (display information) based on the prediction results of the prediction processing unit 134 or the analysis results of the analysis processing unit 135 via the network NW1, and displays it, for example, on the user terminal 14.

[0100] The output control unit 136 outputs, for example, display information as shown in Figures 9 to 11 via the network NW1, and outputs it to, for example, the user terminal 14, causing it to be displayed on the display unit of the user terminal 14.

[0101] Next, the operation of the business support system 100 according to this embodiment will be described with reference to the drawings. Figure 12 is a flowchart showing an example of machine learning processing in the business support system 100 according to this embodiment.

[0102] As shown in Figure 12, the business support server 10 of the business support system 100 first collects training data (step S201). The data collection unit 131 of the business support server 10 collects training data from, for example, the power transmission and distribution business system 15 via the network communication unit 11. The data collection unit 131 collects, for example, past performance information and stores it in the past performance storage unit 124. The data collection unit 131 also collects, for example, linked data and stores it in the linked data storage unit 123.

[0103] Next, the learning data generation unit 132 of the business support server 10 generates a training dataset (training data) from the data acquired by the data collection unit 131 (step S202). The learning data generation unit 132 generates a training dataset, for example, as shown in the training data in Figures 4 to 7. The machine learning unit 133 stores the generated dataset in the learning data storage unit 121.

[0104] Next, the machine learning unit 133 of the business support server 10 executes machine learning processing using the training dataset and generates a trained model (step S203). Based on the training dataset (training data) stored in the training data storage unit 121, the machine learning unit 133 generates trained models corresponding to each business, such as those shown in Figures 4 to 7 (trained model for contract registration / modification processing, trained model for power consumption processing, trained model for various charge calculation processing, and trained model for business invoice processing).

[0105] Next, the machine learning unit 133 stores the trained model in the model storage unit 122 (step S204). For example, the machine learning unit 133 stores the trained model corresponding to each generated task in the model storage unit 122, associating it with task identification information that identifies each task. After the processing in step S204, the machine learning unit 133 terminates its processing.

[0106] Next, with reference to Figure 13, an example of the evaluation process for the business flow of the business support system 100 according to this embodiment will be described. Figure 13 is a flowchart showing an example of the evaluation process of the business flow of the business support system 100 according to this embodiment.

[0107] As shown in Figure 13, the business support server 10 of the business support system 100 first sets the business process and business flow, and whether or not options are available (step S301). The server control unit 13 of the business support server 10 sets the business process and business flow to be evaluated (predicted) and whether or not options are available, based on input information obtained from the user terminal 14, for example via the NW communication unit 11.

[0108] Next, the server control unit 13 determines whether the option is set (step S302). The server control unit 13 checks the option setting and proceeds to step S303 if the option is set (step S302: YES). If the option is not set (not set) (step S302: NO), the server control unit 13 proceeds to step S304.

[0109] In step S303, the server control unit 13 sets the number of workers as a constant. For example, the server control unit 13 sets the number of workers to a pre-set standard value. After processing in step S303, the server control unit 13 proceeds to step S305.

[0110] Furthermore, in step S304, the server control unit 13 sets the worker as a variable. The server control unit 13 sets the number of workers, for example, to be changeable as an evaluation condition. After processing in step S304, the server control unit 13 proceeds to step S305.

[0111] In step S305, the server control unit 13 sets evaluation conditions. The server control unit 13 sets evaluation conditions based on input information obtained from the user terminal 14 via the NW communication unit 11, for example. The server control unit 13 may also select the evaluation conditions from evaluation condition information stored in advance by the evaluation condition storage unit 126.

[0112] Next, the prediction processing unit 134 of the server control unit 13 performs prediction processing using the trained model (step S306). As shown in Figure 8, for example, the prediction processing unit 134 takes evaluation conditions and business flow as input data and predicts, based on the trained model, things like estimated cost, appropriate number of workers, estimated work time, appropriate skill level, etc.

[0113] Next, the analysis processing unit 135 of the server control unit 13 determines whether the objective has been achieved (step S307). The analysis processing unit 135 may determine whether the objective has been achieved based on target information set in advance as evaluation conditions, or it may determine whether the objective has been achieved based on user determination information obtained from the user terminal 14 via the NW communication unit 11. If the objective has been achieved (step S307: YES), the analysis processing unit 135 terminates the process. If the objective has not been achieved (step S307: NO), the analysis processing unit 135 proceeds to step S308.

[0114] In step S308, the analysis processing unit 135 performs analysis and evaluation. The analysis processing unit 135 performs comparison and analysis processing based on the prediction results of the prediction processing unit 134, and outputs comparison display information (display screens G1 to G3) as shown in Figures 9 to 11 to the output control unit 136. The output control unit 136 displays the comparison display information (display screens G1 to G3) on the user terminal 14 via the network communication unit 11, for example.

[0115] Next, the server control unit 13 determines whether or not to modify the business flow (step S309). The server control unit 13 determines whether or not to modify the business flow based on input information obtained from the user terminal 14 via the NW communication unit 11, for example. If the business flow is to be modified (step S309: YES), the server control unit 13 proceeds to step S310. If the business flow is not to be modified (step S309: NO), the server control unit 13 proceeds to step S311.

[0116] In step S310, the server control unit 13 modifies the business flow. The server control unit 13 modifies and changes the business flow based on the modification information obtained from the user terminal 14 via the NW communication unit 11. For example, the server control unit 13 may modify and change a business flow by selecting a different business flow from among the multiple business flows stored in the business flow storage unit 125. After processing in step S310, the server control unit 13 returns to step S302 and executes a re-evaluation process with the modified business flow.

[0117] Furthermore, in step S311, the server control unit 13 performs an evaluation of the proposed content. The analysis processing unit 135 examines the validity of the content of the system proposed parameters (parameters other than the business flow), for example, and optimizes (adjusts) them as necessary. After processing in step S311, the server control unit 13 returns to step S302 and performs a re-evaluation process.

[0118] Next, a modified example of the business support system 100 according to this embodiment will be described with reference to Figures 14 to 19. Figure 14 shows an example of the display of the prediction processing results in the first modified example of this embodiment.

[0119] In the first modified example shown in Figure 14, the work hours are, for example, the daily work hours, and the prediction processing unit 134 predicts the daily work hours for the target business when a business flow corresponding to the target business is input.

[0120] In this case, the output control unit 136 outputs display information such as the display screen G4 shown in Figure 14. As shown in the display screen G4, in this modified example, the output control unit 136 outputs the number of workers and working hours (an example of work hours and work period) for each day.

[0121] Figure 15 is a diagram showing an example of the display of the prediction processing results in a second modified example of this embodiment. In the second modified example shown in Figure 15, the calculation period information includes the meter reading date for electricity consumption, and the machine learning unit 133 performs machine learning based on a dataset that includes the business flow, the work hours or work period for the business, and the meter reading date. When the prediction processing unit 134 receives the business flow and meter reading date corresponding to the business to be predicted (for example, information on schedule A for the meter reading date), it predicts the work hours or work period for the business to be predicted based on the trained model.

[0122] In this case, the output control unit 136 outputs display information such as the display screen G5 shown in Figure 15. As shown in the display screen G5, in this modified example, the output control unit 136 outputs the work hours and work period according to the meter reading date information (for example, the information for schedule A on the meter reading date).

[0123] Figure 16 is a diagram showing an example of the display of the prediction processing results in a third modified example of this embodiment. In the third modified example shown in Figure 16, the calculation period information includes the fee adjustment schedule, and the machine learning unit 133 performs machine learning based on a dataset that includes the business flow, the work effort or work period of the business, and the fee adjustment schedule. When the prediction processing unit 134 receives the business flow and fee adjustment schedule corresponding to the business to be predicted (for example, information for Schedule B), it predicts the work effort or work period of the business to be predicted based on the trained model.

[0124] In this case, the output control unit 136 outputs display information such as the display screen G6 shown in Figure 16. As shown in the display screen G6, in this modified example, the output control unit 136 outputs the work hours and work period according to the fee adjustment schedule (for example, information for schedule B).

[0125] Figure 17 is a diagram showing an example of the display of the prediction processing results in a fourth modified example of this embodiment. In the fourth modified example shown in Figure 17, the machine learning unit 133 further performs machine learning based on a dataset that includes the skills required for the work and the work effort for each skill. When a business flow corresponding to the business to be predicted is input, the prediction processing unit 134 predicts the work effort or work period for each skill required for the business to be predicted, based on the trained model.

[0126] In this case, the output control unit 136 outputs display information such as the display screen G7 shown in Figure 17. As shown in the display screen G7, in this modified example, the output control unit 136 outputs the work hours and work period for each worker's skill level (e.g., worker's skill level and proficiency).

[0127] Figure 18 is a diagram showing an example of the display of the prediction processing results in a fifth modified example of this embodiment. In the fifth modified example shown in Figure 18, the machine learning unit 133 performs machine learning based on a dataset for each power transmission and distribution region or group. Here, the power transmission and distribution region or group is, for example, a region or group on the same meter reading date. When the prediction processing unit 134 receives input for a region or group and a business flow corresponding to the predicted business, it predicts the work effort or work period for the predicted business in that region or group based on the trained model.

[0128] In this case, the output control unit 136 outputs display information such as the display screen G8 shown in Figure 18. As shown in the display screen G8, in this modified example, the output control unit 136 outputs the work hours and work period for each region (or group).

[0129] Figure 19 is a diagram showing an example of the display of the prediction processing results in the sixth modified example of this embodiment. In the sixth modified example shown in Figure 19, the machine learning unit 133 performs machine learning based on a dataset that includes information indicating an irregular event and the work effort or work period for the business in the event that the irregular event occurs. When information indicating an irregular event is input, the prediction processing unit 134 predicts the work effort or work period for the business in the event that the irregular event occurs, based on the trained model.

[0130] In this case, the output control unit 136 outputs display information such as the display screen G9 shown in Figure 19. As shown in the display screen G9, in this modified example, the output control unit 136 outputs the work hours and work period corresponding to the irregular event.

[0131] In addition, an irregular event is, for example, a power outage, and the machine learning unit 133 may further perform machine learning based on the power outage information. In this case, when power outage information is input, the prediction processing unit 134 may predict the amount of work or the duration of work in the event of a power outage based on the trained model.

[0132] Furthermore, an irregular event is a deficiency in power data, and the machine learning unit 133 may perform machine learning based on the information about the power data deficiency. In this case, when the prediction processing unit 134 receives information about the power data deficiency, it may predict the amount of work required or the duration of work in the event of a power data deficiency based on the trained model.

[0133] Furthermore, an irregular event is a failure in the coordination of power data between power-related businesses, and the machine learning unit 133 may perform machine learning based on the information regarding the failure in power data coordination. In this case, when the prediction processing unit 134 receives information regarding the failure in power data coordination, it may predict the amount of work required or the duration of work in the event of a failure in power data coordination between power-related businesses based on the trained model.

[0134] Furthermore, if an irregular event is a deficiency in the contract data related to electricity, the machine learning unit 133 may perform machine learning based on the information about the deficiency in the contract data. In this case, if the prediction processing unit 134 receives information about the deficiency in the contract data, it may predict the amount of work required or the duration of work in the event of a deficiency in the contract data based on the trained model.

[0135] As described above, the business support system 100 according to this embodiment comprises a machine learning unit 133 and a prediction processing unit 134. The machine learning unit 133 performs machine learning based on a dataset that includes a business flow containing multiple tasks (for example, a business flow for business processing GS, an example of business requirements) and the work effort or work period for each task, and generates a trained model. When a business flow that is expected to change is input, the prediction processing unit 134 predicts at least the work effort or work period for each task included in the business flow based on the trained model.

[0136] As a result, the business support system 100 according to this embodiment performs machine learning based on a dataset that includes business flows (business requirements) and the work effort or work period of each business, and predicts the work effort or work period for each business. Therefore, for example, when changing or improving a business system, it can accurately predict the work effort or work time from the business flows (business requirements). Thus, the business support system 100 according to this embodiment can accurately evaluate the business system at the business flow (business requirements) stage.

[0137] Furthermore, in the business support system 100 according to this embodiment, the machine learning unit 133 performs machine learning based on a dataset that includes a business flow, the work hours or work period of the business, and calculation period information, which is time information necessary for calculating electricity charges, to generate a trained model. In addition, when the business flow of a business to be predicted is input, the prediction processing unit 134 predicts the work hours or work period of the business to be predicted based on the trained model.

[0138] As a result, the business support system 100 according to this embodiment performs machine learning based on a dataset including calculation period information to predict the work hours or work period of the target business. Therefore, for example, in business related to electricity charge calculation, it can predict work hours or work time more accurately. Thus, the business support system 100 according to this embodiment can predict work hours or work time more accurately from the business flow (business requirements), and can accurately evaluate the business system at the business flow (business requirements) stage.

[0139] In this embodiment, the work hours are the work hours per day, and the prediction processing unit 134 predicts the daily work hours for the target business when a business flow corresponding to the target business is input.

[0140] As a result, the business support system 100 according to this embodiment can predict the daily workload of the target business and appropriately evaluate the business system at the business flow (business requirements) stage.

[0141] Furthermore, in this embodiment, the calculation period information includes the meter reading date for electricity consumption, and the machine learning unit 133 performs machine learning based on a dataset that includes the business flow, the work hours or work period of the business, and the meter reading date. In other words, the machine learning unit 133 further performs machine learning based on a dataset that includes the meter reading date for electricity consumption. When the prediction processing unit 134 receives the business flow and meter reading date corresponding to the business to be predicted, it predicts the work hours or work period of the business to be predicted based on the trained model.

[0142] As a result, the business support system 100 according to this embodiment can predict the amount of work required or the duration of work, taking into account the meter reading schedule, and can further appropriately evaluate the business system at the business flow (business requirements) stage.

[0143] Furthermore, in this embodiment, the calculation period information includes the fee adjustment schedule, and the machine learning unit 133 performs machine learning based on a dataset that includes the business flow, the work hours or work period of the business, and the fee adjustment schedule. That is, the machine learning unit 133 further performs machine learning based on a dataset that includes the fee adjustment schedule. When the prediction processing unit 134 receives the business flow and fee adjustment schedule corresponding to the business to be predicted as input, it predicts the work hours or work period of the business to be predicted based on the trained model.

[0144] As a result, the business support system 100 according to this embodiment can predict the amount of work or the duration of work while taking into account the fee adjustment schedule, and can further appropriately evaluate the business system at the business flow (business requirements) stage.

[0145] In this embodiment, the machine learning unit 133 further performs machine learning based on a dataset that includes the skills required for the work and the work effort for each skill. When a business flow corresponding to the business to be predicted is input, the prediction processing unit 134 predicts the work effort or work period for each skill required for the business to be predicted, based on the trained model.

[0146] As a result, the business support system 100 according to this embodiment can predict the amount of work required or the duration of work, taking into account the skills necessary for the worker's tasks, and can more appropriately evaluate the business system at the business flow (business requirements) stage.

[0147] Furthermore, the business support system 100 according to this embodiment includes a data collection unit 131 (collection unit) and a learning data generation unit 132. The data collection unit 131 collects, for each worker, business identification information that identifies the business and historical information (e.g., past performance) of work hours or work period. The learning data generation unit 132 generates a dataset by aggregating the work hours or work period for each business indicated by the business identification information based on the historical information (e.g., past performance).

[0148] As a result, the business support system 100 according to this embodiment can, for example, collect the latest data for each worker when evaluating a business system, generate a dataset of training data, and use it for machine learning, thereby enabling more accurate prediction of work effort or work time.

[0149] In this embodiment, the machine learning unit 133 performs machine learning based on a dataset for each power transmission and distribution region or group. When a region or group and a business flow corresponding to the predicted business are input, the prediction processing unit 134 predicts the work effort or work period for the predicted business in that region or group based on the trained model.

[0150] As a result, the business support system 100 according to this embodiment can predict the amount of work or the duration of the target business for each region or group corresponding to the meter reading date, for example, when business processing is carried out by distributing meter reading dates based on daily charges, and the business system can be evaluated more appropriately.

[0151] In this embodiment, the machine learning unit 133 performs machine learning based on a dataset that includes a business flow corresponding to the invoice creation work related to transmission charges (for example, the business invoice processing work for businesses) and the work hours or work period for the work. When the prediction processing unit 134 receives a business flow corresponding to the invoice creation work to be predicted, it predicts the work hours or work period for the invoice creation work based on the trained model.

[0152] As a result, the business support system 100 according to this embodiment can appropriately evaluate the invoice creation process related to transmission charges (for example, the process of processing invoices for businesses) at the business flow (business requirements) stage.

[0153] In this embodiment, the machine learning unit 133 performs machine learning based on a dataset that includes the workflow of electricity billing operations (for example, operations for various billing calculation processes) and the work hours or duration of the operations. When the prediction processing unit 134 receives a workflow corresponding to the billing operation to be predicted, it predicts the work hours or duration of the billing operation based on the trained model.

[0154] As a result, the business support system 100 according to this embodiment can appropriately evaluate the tasks of electricity billing (for example, tasks of various billing calculation processes) at the business flow (business requirements) stage.

[0155] In this embodiment, the machine learning unit 133 performs machine learning based on a dataset that includes information indicating an irregular event and the amount of work or duration required for the work if the irregular event occurs. When information indicating an irregular event is input, the prediction processing unit 134 predicts the amount of work or duration required for the work if the irregular event occurs, based on the trained model.

[0156] As a result, the business support system 100 according to this embodiment can accurately predict the workload or duration of work in the event of an irregular event, and can evaluate the business system taking into account the occurrence of an irregular event.

[0157] In this embodiment, the irregular event is a power outage, and the machine learning unit 133 further performs machine learning based on the power outage information. When power outage information is input, the prediction processing unit 134 predicts the workload or duration of the work in the event of a power outage based on the trained model.

[0158] As a result, the business support system 100 according to this embodiment can accurately predict the amount of work required or the duration of work in the event of a power outage, and can evaluate the business system more appropriately, taking into account the possibility of a power outage.

[0159] In this embodiment, the irregular event is a deficiency in power data. The machine learning unit 133 further performs machine learning based on the information about the power data deficiency. When the prediction processing unit 134 receives information about the power data deficiency, it predicts the workload or duration of the work that would occur if the power data deficiency were to occur, based on the trained model.

[0160] As a result, the business support system 100 according to this embodiment can accurately predict the workload or duration of operations in the event of deficiencies in power data, and can more appropriately evaluate the business system in consideration of the possibility of deficiencies in power data.

[0161] Furthermore, in this embodiment, the irregular event is a failure in the coordination of power data between power-related businesses. The machine learning unit 133 further performs machine learning based on the information about the power data coordination failure. When the prediction processing unit 134 receives information about the power data coordination failure, it predicts the amount of work or the duration of work that would be required if a power data coordination failure occurred between power-related businesses, based on the trained model.

[0162] As a result, the business support system 100 according to this embodiment can accurately predict the workload or duration of work in the event of a failure in the linkage of power data between power-related businesses, and can more appropriately evaluate the business system in consideration of the possibility of a failure in the linkage of power data between power-related businesses.

[0163] In this embodiment, the irregular event is a deficiency in the contract data related to electricity. The machine learning unit 133 further performs machine learning based on the information about the deficiency in the contract data. When the prediction processing unit 134 receives information about the deficiency in the contract data, it predicts the amount of work required or the duration of work in the event of a deficiency in the contract data based on the trained model.

[0164] As a result, the business support system 100 according to this embodiment can accurately predict the workload or duration of work in the event of deficiencies in electricity contract data, and can more appropriately evaluate the business system in consideration of the possibility of deficiencies in electricity contract data.

[0165] Furthermore, the business support system 100 according to this embodiment includes an output control unit 136. The output control unit 136 changes the setting conditions, including the business flow, and compares the predicted work hours or work period of the business predicted by the prediction processing unit 134, and outputs the result.

[0166] As a result, the business support system 100 according to this embodiment can compare prediction results with changed setting conditions, including the business flow, and thus can evaluate the business system more appropriately at the business flow (business requirements) stage.

[0167] Furthermore, in this embodiment, the output control unit 136 compares and outputs the work effort or work period for multiple modified business flows.

[0168] As a result, the business support system 100 according to this embodiment can compare prediction results for multiple modified business flows, thereby enabling a more appropriate evaluation of the business system at the business flow (business requirements) stage.

[0169] Furthermore, in this embodiment, the output control unit 136 compares the work hours or work period for the modified business flow with the work hours or work period for a similar business flow that is similar to the modified business flow, and outputs the result.

[0170] As a result, the business support system 100 according to this embodiment can compare the prediction results for the modified business flow with similar business flows, thereby enabling a more appropriate evaluation of the business system at the business flow (business requirements) stage.

[0171] Furthermore, in this embodiment, the output control unit 136 compares and outputs the work effort or work period for multiple modified business flows. As a result, the business support system 100 according to this embodiment can compare prediction results for the business flow after the number of changes, thereby enabling a more appropriate evaluation of the business system at the business flow (business requirements) stage.

[0172] Furthermore, the business support method according to this embodiment includes a machine learning step and a prediction processing step. In the machine learning step, the machine learning unit 133 performs machine learning based on a dataset that includes a business flow containing multiple tasks and the work effort or work period of the tasks to generate a trained model. In the prediction processing step, when a business flow that is expected to change is input, the prediction processing unit 134 predicts at least the work effort or work period for each task included in the business flow based on the trained model.

[0173] As a result, the business support method according to this embodiment achieves the same effects as the business support system 100 described above. The business support method according to this embodiment can, for example, accurately predict the amount of work or working time from the business flow (business requirements) when changing or improving a business system, and can accurately evaluate the business system at the business flow (business requirements) stage.

[0174] Furthermore, the business support method according to this embodiment includes a machine learning step and a prediction processing step. In the machine learning step, the machine learning unit 133 performs machine learning based on a dataset including a business flow, the work hours or work period of the business, and calculation period information which is time information necessary for calculating electricity charges, and generates a trained model. In the prediction processing step, when the business flow of the business to be predicted is input, the prediction processing unit 134 outputs the work hours or work period of the business to be predicted based on the trained model.

[0175] As a result, the business support method according to this embodiment achieves the same effects as the business support system 100 described above. The business support method according to this embodiment can predict the work hours or man-hours more accurately in tasks related to electricity bill calculation, and can accurately evaluate the business system at the business flow (business requirements) stage.

[0176] Figure 20 illustrates an example of the hardware configuration of the business support server 10 according to this embodiment. As shown in Figure 20, the business support server 10 includes a communication device H11, memory H12, and a processor H13.

[0177] Communication device H11 is a communication device that can connect to network NW1, such as a LAN card. Memory H12 is a storage device such as RAM, flash memory, or HDD, and stores various information and programs used by the business support server 10.

[0178] Processor H13 is a processing circuit that includes, for example, a CPU. Processor H13 executes various processes of the business support server 10 by running programs stored in memory H12.

[0179] This disclosure is not limited to the embodiments described above and may be modified without departing from the spirit of this disclosure. For example, in the above embodiment, the business support server 10 was described as being implemented as a single server device, but it is not limited to this, and may be implemented as multiple devices, such as multiple server devices.

[0180] Furthermore, while the above embodiments have described examples where the business processes are contract registration / modification processing, electricity usage processing, various fee calculation processing, and invoice processing for businesses, the disclosure is not limited to these, and may be applied to other business processes.

[0181] Furthermore, each component of the business support system 100 described above has a computer system inside. The processing in each component of the business support system 100 may be performed by recording a program for realizing the functions of each component of the business support system 100 onto a computer-readable recording medium, loading the program recorded on this recording medium into the computer system, and executing it. Here, "loading the program recorded on the recording medium into the computer system and executing it" includes installing the program into the computer system. Here, "computer system" includes the OS and hardware such as peripheral devices.

[0182] Furthermore, "computer system" may include multiple computer devices connected via a network, including communication lines such as the Internet, WAN, LAN, and dedicated lines. Also, "computer-readable recording medium" refers to portable media such as flexible disks, magneto-optical disks, ROMs, and CD-ROMs, as well as storage devices such as hard disks built into the computer system. Thus, the recording medium storing the program may also be a non-transient recording medium such as a CD-ROM.

[0183] Furthermore, the recording medium also includes internal or external recording media accessible from the distribution server for distributing the program. The program may be divided into multiple parts, downloaded at different times, and then combined in each configuration of the business support system 100, or each divided program may be distributed by a different distribution server. Additionally, "computer-readable recording medium" includes volatile memory (RAM) within a computer system that acts as a server or client when the program is transmitted over a network, which retains the program for a certain period of time. Moreover, the program may be intended to implement only a part of the functions described above. Furthermore, the program may be a so-called differential file (differential program) that can implement the functions described above in combination with a program already recorded in the computer system. [Explanation of Symbols]

[0184] 1...Transmission and distribution company, 2...Electricity consumer, 3...Power generation company, 4...Electricity retail company, 10...Business support server, 11...NW communication unit, 12...Server memory unit, 13...Server control unit, 14...User terminal, 15...Transmission and distribution business system, 20...Electricity meter, 21...Meter reading terminal, 30...Power generation business system, 40...Electricity retail business system, 100...Business support system, 121...Learning data memory unit, 122...Model memory unit, 123...Linked data memory unit, 124...Past performance memory unit, 125...Business flow memory unit, 126...Evaluation condition memory unit, 127...Proposal content memory unit, 128...Analysis result memory unit, 131...Data collection unit, 132...Learning data generation unit, 133...Machine learning unit, 134...Prediction processing unit, 135...Analysis processing unit, 136...Output control unit, NW1...Network

Claims

1. A machine learning unit that performs machine learning based on a dataset including business requirements that include multiple tasks and the work hours or work period of the said tasks to generate a trained model, When the aforementioned business requirements that are expected to change are input, a prediction processing unit predicts at least the amount of work effort or the duration of work for each of the aforementioned tasks included in the aforementioned business requirements, based on the trained model. A business support system equipped with the following features.

2. The aforementioned work hours are the work hours per day. The prediction processing unit predicts the daily workload for the target task when the business requirements corresponding to the target task are input. The business support system according to claim 1.

3. The aforementioned machine learning unit further performs machine learning based on a dataset that includes the meter reading date for electricity consumption. When the forecasting processing unit receives the business requirements corresponding to the business to be predicted and the meter reading date, it predicts the work hours or work period of the business to be predicted based on the trained model. The business support system according to claim 1.

4. The aforementioned machine learning unit further performs machine learning based on a dataset including the fee adjustment schedule. When the forecasting processing unit receives the business requirements and fee adjustment schedule corresponding to the business to be forecasted, it forecasts the work hours or work period of the business to be forecasted based on the trained model. The business support system according to claim 1.

5. The aforementioned machine learning unit further performs machine learning based on a dataset that includes the skills required for the work and the work effort for each of those skills. When the aforementioned prediction processing unit receives the business requirements corresponding to the business to be predicted, it predicts the work effort or work period for each skill required for the business to be predicted, based on the trained model. The business support system according to claim 1.

6. A collection unit collects, for each worker, task identification information to identify the aforementioned task and historical information such as work hours or work period. Based on the aforementioned history information, a learning data generation unit aggregates the work hours or work period for each of the tasks indicated by the task identification information and generates the aforementioned dataset. The business support system according to claim 1, comprising:

7. The machine learning unit performs machine learning based on the dataset for each power transmission and distribution region or group. When the region or group and the business requirements corresponding to the business to be predicted are input to the prediction processing unit, it predicts the man-hours or work period of the business to be predicted for that region or group based on the trained model. The business support system according to claim 1.

8. The machine learning unit performs machine learning based on the dataset which includes the business requirements for the invoice creation work related to transmission charges and the work hours or work period for the said work. When the prediction processing unit receives the business requirements corresponding to the invoice creation task to be predicted, it predicts the work hours or work period for the invoice creation task based on the trained model. The business support system according to claim 1.

9. The machine learning unit performs machine learning based on the dataset, which includes the business requirements for electricity rate calculation operations and the work hours or work period for those operations. When the aforementioned prediction processing unit receives the business requirements corresponding to the fee calculation business to be predicted, it predicts the work hours or work period of the fee calculation business based on the trained model. The business support system according to claim 1.

10. The machine learning unit performs machine learning based on the dataset which includes information indicating an irregular event and the work effort or work period of the task when the irregular event occurs. When the prediction processing unit receives information indicating the irregular event, it predicts the workload or duration of the work if the irregular event occurs, based on the trained model. The business support system according to claim 1.

11. The aforementioned irregular event was a power outage. The aforementioned machine learning unit further performs machine learning based on the power outage information. When the power outage information is input, the prediction processing unit predicts the workload or duration of the task in the event of a power outage based on the trained model. The business support system according to claim 10.

12. The aforementioned irregular event was due to a deficiency in power data. The machine learning unit further performs machine learning based on the information regarding the deficiencies in the power data, When the prediction processing unit receives information about deficiencies in the power data, it predicts the workload or duration of the task if deficiencies in the power data occur, based on the trained model. The business support system according to claim 10.

13. The aforementioned irregular incident was due to a failure in the sharing of power data between power-related businesses. The aforementioned machine learning unit further performs machine learning based on the information regarding the discrepancies in the power data linkage, When information regarding a discrepancy in the linkage of power data is input, the prediction processing unit predicts, based on the trained model, the amount of work required or the duration of work in the event of a discrepancy in the linkage of power data between power-related businesses. The business support system according to claim 10.

14. The aforementioned irregular event was due to deficiencies in the contract data regarding electricity. The aforementioned machine learning unit further performs machine learning based on the information regarding deficiencies in the contract data, When information regarding deficiencies in the contract data is input, the prediction processing unit predicts, based on the trained model, the amount of work required or the duration of the work if deficiencies in the contract data occur. The business support system according to claim 10.

15. The system includes an output control unit that modifies the setting conditions, including the aforementioned business requirements, and compares them with the work hours or work period predicted by the prediction processing unit for the aforementioned business, and outputs the result. A business support system according to any one of claims 1 to 14.

16. The output control unit compares the work hours or work period of the business with respect to the business requirements before the change and the work hours or work period of the business with respect to the business requirements after the change, and outputs the result. The business support system according to claim 15.

17. The output control unit compares the work hours or duration of the task for the modified task requirements with the work hours or duration of the task for similar task requirements that are similar to the modified task requirements, and outputs the result. The business support system according to claim 15.

18. The output control unit compares and outputs the work effort or work period of the task for multiple modified business requirements. The business support system according to claim 15.

19. The machine learning department performs machine learning based on a dataset that includes business requirements encompassing multiple tasks and the work effort or duration of those tasks, to generate a trained model. When the prediction processing unit receives the business requirements that are expected to change, it predicts, based on the trained model, at least the amount of work time or the duration of work for each of the tasks included in the business requirements. Business support methods.

20. On the computer, A machine learning step that generates a trained model by performing machine learning based on a dataset that includes business requirements encompassing multiple tasks and the work effort or work period for said tasks, When the business requirements that are expected to change are entered, a prediction processing step is performed to predict at least the amount of work or the duration of work for each of the tasks included in the business requirements, based on the trained model. A program to execute.

Images