Operation support system, operation support method, and operation support program

Abstract

A prediction unit (121) is input with target operation activity data and uses a prediction model to predict an evaluation index for a target operation activity, the target operation activity data including target activity specifying data that indicates information specifying the target operation activity carried out during a target operation, and target activity content data that indicates the content of the target operation activity.

Description

Business support system, business support method, and business support program

[0001] The present disclosure relates to predicting performance metrics for business activities.

[0002] Patent Document 1 discloses a technology for appropriately supporting sales activities by predicting actions to be taken from the present onward in order to increase the likelihood of receiving orders.

[0003] Patent Literature 1 exemplifies many types of actions that sales representatives take toward customers, such as "seminars," "emails," "visits," and "websites," as sales activities used in predicting the probability of success. However, only the "time series of sales activities," "customer attributes," and "order success or failure" are used to predict the probability of success of sales activities, and does not take into account the detailed content of each sales activity (text such as business trip reports and minutes of meetings). Furthermore, no consideration is given to customer reactions to each sales activity (positive opinions, negative opinions, new requests, etc.).

[0004] International Publication No. 2021 / 192197

[0005] The present disclosure aims to enable evaluation indicators for business activities to be predicted with higher accuracy.

[0006] The business support system of the present disclosure includes a prediction unit that inputs target business activity data including target activity identification data indicating information identifying a target business activity carried out in a target business and target activity content data indicating the content of the target business activity, and predicts evaluation indicators for the target business activity using a prediction model.

[0007] According to the present disclosure, not only information specifying a business activity, such as the type of business activity, but also detailed content of the business activity is taken into consideration, thereby making it possible to predict evaluation indicators for the business activity with higher accuracy.

[0008] 1 is a diagram illustrating a configuration of a business support system 100 according to a first embodiment. A functional configuration diagram of the business support system 100 according to the first embodiment. A flowchart of a business support method according to the first embodiment. A functional configuration diagram of a sales support system 100A according to the first embodiment. A functional configuration diagram of a production site support system 100B according to the first embodiment. A diagram showing an overview of the sales support system 100A according to the first embodiment. A functional configuration diagram of the business support system 100 according to the second embodiment. A flowchart of a business support method according to the second embodiment. A functional configuration diagram of the sales support system 100A according to the second embodiment. A functional configuration diagram of the production site support system 100B according to the second embodiment. A diagram showing an overview of the sales support system 100A according to the second embodiment. A functional configuration diagram of the business support system 100 according to the third embodiment. A functional configuration diagram of the business support system 100 according to the third embodiment. A flowchart of a business support method according to the third embodiment. A functional configuration diagram of the sales support system 100A according to the third embodiment. A functional configuration diagram of the production site support system 100B according to the third embodiment. FIG. 1 is a hardware configuration diagram of a task support system 100 according to an embodiment.

[0009] In the embodiments and drawings, the same or corresponding elements are denoted by the same reference numerals. The description of elements denoted by the same reference numerals as those already described will be omitted or simplified as appropriate. Arrows in the drawings primarily indicate the flow of data or the flow of processing.

[0010] First Embodiment A business support system 100 will be described with reference to FIGS.

[0011] ***Description of Configuration*** The configuration of the business support system 100 will be described with reference to Fig. 1. The business support system 100 can be configured with one device. However, the business support system 100 may also be configured with multiple devices.

[0012] The business support system 100 is a computer that includes hardware such as a processor 101, a memory 102, an auxiliary storage device 103, a communication device 104, and an input / output interface 105. These pieces of hardware are connected to each other via signal lines.

[0013] The processor 101 is an IC that performs arithmetic processing and controls other hardware. For example, the processor 101 is a CPU, a DSP, or a GPU. IC is an abbreviation for Integrated Circuit. CPU is an abbreviation for Central Processing Unit.

[0014] The memory 102 is a volatile or non-volatile storage device. The memory 102 is also called a primary storage device or a main memory. For example, the memory 102 is a RAM. Data stored in the memory 102 is saved in the secondary storage device 103 as needed. RAM is an abbreviation for Random Access Memory.

[0015] The auxiliary storage device 103 is a non-volatile storage device. For example, the auxiliary storage device 103 is a ROM, a HDD, a flash memory, or a combination of these. Data stored in the auxiliary storage device 103 is loaded into the memory 102 as needed. ROM is an abbreviation for Read Only Memory. HDD is an abbreviation for Hard Disk Drive.

[0016] The communication device 104 is a receiver and a transmitter. For example, the communication device 104 is a communication chip or a NIC. Communication in the business support system 100 is performed using the communication device 104. NIC is an abbreviation for Network Interface Card.

[0017] The input / output interface 105 is a port to which an input device and an output device are connected. For example, the input / output interface 105 is a USB terminal, the input devices are a keyboard and a mouse, and the output device is a display. Input and output of the business support system 100 are performed via the input / output interface 105. USB is an abbreviation for Universal Serial Bus.

[0018] The business support system 100 includes elements such as a prediction model creation unit 111 and a prediction unit 121. These elements are realized by software.

[0019] The auxiliary storage device 103 stores a business support program for causing the computer to function as the prediction model creation unit 111 and the prediction unit 121. The business support program is loaded into the memory 102 and executed by the processor 101. The auxiliary storage device 103 also stores an OS. At least a portion of the OS is loaded into the memory 102 and executed by the processor 101. The processor 101 executes the business support program while running the OS. OS is an abbreviation for Operating System.

[0020] Data (input data, output data, etc.) of the business support program is stored in the storage unit 190. The memory 102 functions as the storage unit 190. However, a storage device such as the auxiliary storage device 103, a register in the processor 101, or a cache memory in the processor 101 may function as the storage unit 190 instead of or together with the memory 102.

[0021] The business support program can be recorded (stored) in a computer-readable manner on a non-volatile recording medium such as an optical disk or a flash memory.

[0022] 2 shows the functional configuration of the business support system 100. The business database 200 has multiple pieces of business data 201. The business data 201 is data on past business operations. Examples of business operations include sales and production. The business data 201 serves as training data during machine learning to create the prediction model 191.

[0023] The business data 201 includes business activity data 210 and business performance data 220 .

[0024] The business activity data 210 includes activity specification data 211 and activity content data 212. The business activity data 210 is business activity data of past business. The activity specification data 211 is activity specification data of past business. The activity content data 212 is activity content data of past business.

[0025] The activity identification data indicates information for identifying a business activity carried out in a business, such as the business name, the customer, the person in charge, the activity date and time, and the activity type.

[0026] The activity content data indicates the content of business activities carried out in a business. For example, the activity content data indicates text information accompanying the business activities. Examples of the activity content data include reports and minutes of meetings.

[0027] The business performance data 220 indicates past business performance. For example, if the business is sales, the business performance data 220 indicates order results, etc., and if the business is production, the business performance data 220 indicates production volume, etc.

[0028] The business activity data 230 is business activity data of a target business. For example, the target business is a current business being performed.

[0029] The business activity data 230 includes activity specification data 231 and activity content data 232. The activity specification data 231 is activity specification data of the target business (target activity specification data). The activity content data 232 is activity content data of the target business (target activity content data).

[0030] The prediction model 191 is a trained model that predicts evaluation indices for business activities using activity specification data and activity content data as input. Examples of evaluation indices include order probability and production efficiency.

[0031] The evaluation index data 192 indicates evaluation indexes for the business activities of the target business (target business activities).

[0032] ***Description of Operation*** The operational procedure of the business support system 100 corresponds to a business support method. Also, the operational procedure of the business support system 100 corresponds to a processing procedure by a business support program.

[0033] The business support method will be described with reference to Fig. 3. In step S110, the prediction model creation unit 111 learns a set of activity specification data 211, activity content data 212, and business performance data 220 for each past business, and creates a prediction model 191.

[0034] In step S120, the prediction unit 121 receives the business activity data 230. The business activity data 230 includes activity specification data 231 and activity content data 232.

[0035] For example, an administrator inputs the business activity data 230 into the business support system 100 , and the prediction unit 121 receives the input business activity data 230 .

[0036] In step S130, the prediction unit 121 receives the business activity data 230 as an input and predicts the evaluation index for the target business activity using the prediction model 191.

[0037] In step S140, the prediction unit 121 outputs the evaluation index data 192. The evaluation index data 192 indicates the evaluation index for the target business activity.

[0038] For example, the prediction unit 121 displays the evaluation index data 192 on a display.

[0039] ***Description of First Embodiment*** A sales support system 100A will be described with reference to FIG. 4. The sales support system 100A is a business support system 100 for supporting sales operations. A sales database 200A is a business database 200 in the sales support system 100A. Sales data 201A is business data 201 in the sales support system 100A. Sales activity data 210A is business activity data 210 in the sales support system 100A. Activity identification data 211A is activity identification data 211 in the sales support system 100A. For example, the activity identification data 211A indicates the case name, customer, sales representative, activity date and time, activity type, etc. Examples of activity types include seminars, emails, visits, and demonstrations at exhibitions. Activity content data 212A is activity content data 212 in the sales support system 100A. Examples of activity content include business trip reports and meeting minutes. The activity content may include seminar information (video footage, voice recognition results, text transcription results of the voice recognition results, etc.). The activity content may also include exhibition panel information (text, drawings, graphs, etc.). The order history data 220A is the business performance data 220 in the sales support system 100A. For example, the order history data 220A indicates the name of the order, the customer, the order date, the order amount, the sales representative, etc. The prediction model creation unit 111 learns the combination of the activity identification data 211A, the activity content data 212A, and the order history data 220A for each past sales transaction to create the prediction model 191. The sales activity data 230A is the business activity data 230 in the sales support system 100A. The activity identification data 231A is the activity identification data 231 in the sales support system 100A and indicates information identifying the sales activity (target sales activity) carried out in the sales of the target project. For example, the activity identification data 231A indicates, for each sales representative (A, B), the project name, customer, sales representative, activity date and time, activity type, etc. The activity content data 232A is the activity content data 232 in the sales support system 100A and indicates the content of the target sales activity. The prediction unit 121 inputs the activity identification data 231A and the activity content data 232A and uses the prediction model 191 to predict the order probability of the target project.The order probability data 192A is the evaluation index data 192 in the sales support system 100A. For example, the order probability data 192A indicates the project name, customer, sales representative, order probability, etc. for each sales representative (A, B). The manager of the sales department refers to the order probability data 192A and gives instructions for sales activities based on the current order probability for each project.

[0040] ***Description of Second Embodiment*** A production site support system 100B will be described with reference to FIG. 5. The production site support system 100B is a task support system 100 for supporting production tasks. A production database 200B is a task database 200 in the production site support system 100B. Production data 201B is task data 201 in the production site support system 100B. Production plan data 210B is task activity data 210 in the production site support system 100B. Activity specification data 211B is activity specification data 211 in the production site support system 100B. For example, the activity specification data 211B indicates the model name, customer, production date and time, planned number of units, worker, activity type, etc. Examples of activity types include assembly, precision measurement, and visual inspection. Activity content data 212B is activity content data 212 in the production site support system 100B. Examples of activity content include a work diary, training participation record, and on-site improvement activity record. The production performance data 220B is the business performance data 220 in the production site support system 100B. For example, the production performance data 220B indicates the model name, customer, production date and time, actual number of units, etc. The prediction model creation unit 111 creates the prediction model 191 by learning the combination of the activity identification data 211B, activity content data 212B, and production performance data 220B for each past production plan. The production plan data 230B is the business activity data 230 in the production site support system 100B. The activity identification data 231B is the activity identification data 231 in the production site support system 100B and indicates information that identifies the production activity carried out in the target production plan (target production activity). For example, the activity identification data 231B indicates, for each worker (X, Y), the model name, customer, production date and time, planned number of units, worker, activity type, etc. The activity content data 232B is the activity content data 232 in the production site support system 100B and indicates the content of the target production activity. The prediction unit 121 inputs the activity specification data 231B and the activity content data 232B for each target production plan and predicts the production efficiency of the target production plan for each worker using the prediction model 191. The production efficiency data 192B is the evaluation index data 192 in the production site support system 100B.For example, the production efficiency data 192B indicates the model name, customer, production date and time, predicted production volume, etc. for each worker (X, Y). The production site manager refers to the production efficiency data 192B and adjusts the production plan based on the production efficiency of each worker for each production plan. Note that multiple production data 201B can be created by appropriately changing the combination of worker and model name. Using multiple production data 201B makes it possible to compare overall production efficiencies.

[0041] ***Features of First Embodiment*** The business support system 100 learns a business evaluation index prediction model based on past business activity data, activity content data accompanying the business activity data, and business evaluation index data. The business evaluation index prediction model predicts business evaluation indexes. The business support system 100 predicts business evaluation indexes for current business activities using the business evaluation index prediction model based on current business activity data and activity content data accompanying the business activity data.

[0042] Figure 6 shows an overview of the sales support system 100A. The sales support system 100A learns an order probability prediction model based on past sales activity data, activity content data accompanying the sales activity data, and order record data. The order probability prediction model predicts the order probability. The sales support system 100A predicts the order probability for the current sales activity using the order probability prediction model based on the current sales activity data and activity content data accompanying the sales activity data. In Figure 6, the <explanatory variables> include the case name, customer, sales representative, activity type, and activity content. The <objective variable> is the order record (order won / lost). The <output> is the order probability (order won: x%, lost (100-x)%).

[0043] The production site support system 100B learns a production efficiency prediction model based on past production plan data, activity content data accompanying the production plan data, and production performance data. The production efficiency prediction model predicts production efficiency. The production site support system 100B predicts production efficiency for the current production plan using the production efficiency prediction model based on the current production plan data and activity content data accompanying the production plan data.

[0044] ***Effects of First Embodiment*** The business support system 100 can improve the prediction accuracy of the evaluation index for the target business by using the activity content data.

[0045] The sales support system 100A predicts the probability of receiving an order based on sales activity data. This makes it possible to predict the probability of receiving an order for each sales representative and each customer. Furthermore, by using activity content data, the accuracy of predictions of the probability of receiving an order can be improved. The manager of the sales department can give appropriate instructions to the sales representative according to the probability of receiving an order for each business negotiation. In addition, the manager can consider measures for cases with a low probability of receiving an order.

[0046] The production site support system 100B predicts production efficiency based on production plan data. This makes it possible to predict production efficiency for each worker. Furthermore, by using the activity content data for each worker, the accuracy of production efficiency predictions can be improved. The production department manager can improve the efficiency of production operations by adjusting the production plan in consideration of differences in production efficiency depending on the combination of worker and machine model.

[0047] Second Embodiment A mode of presenting data on tasks similar to a target task will be described below, focusing mainly on the differences from the first embodiment, with reference to FIGS.

[0048] ***Description of Configuration*** The configuration of the business support system 100 will be described with reference to Fig. 7. The business support system 100 further includes a detection model creation unit 112 and a detection unit 122. The business support program further causes a computer to function as the detection model creation unit 112 and the detection unit 122.

[0049] 8 shows the functional configuration of the task support system 100. The detection model 193 is a trained model that receives activity specification data and activity content data as input and detects similar task data.

[0050] The similar business activity data 194 is business data of a business similar to the target business (similar business).

[0051] ***Description of Operation*** The business support method will be described with reference to Fig. 9. In step S210, the prediction model creation unit 111 learns the set of activity identification data 211, activity content data 212, and business performance data 220 for each past business, and creates the prediction model 191. Step S210 is the same as step S110 in the first embodiment.

[0052] In step S220, the detection model creation unit 112 learns the pair of activity specification data 211 and activity content data 212 for each past task, and creates the detection model 193.

[0053] In step S230, the prediction unit 121 receives the business activity data 230. The business activity data 230 includes activity specification data 231 and activity content data 232. Step S230 is the same as step S120 in the first embodiment.

[0054] In step S240, the prediction unit 121 receives the business activity data 230 as an input and predicts the evaluation index for the target business using the prediction model 191. Step S240 is the same as step S130 in the first embodiment.

[0055] In step S250, the prediction unit 121 outputs the evaluation index data 192. Step S250 is the same as step S140 in the first embodiment.

[0056] In step S260, the detection unit 122 receives the work activity data 230 as input and uses the detection model 193 to detect work activity data 210 of similar work from among the multiple work activity data 210. The detected work activity data 210 is referred to as similar work activity data 194.

[0057] In step S270, the detection unit 122 outputs the similar work activity data 194.

[0058] For example, the detection unit 122 displays the similar work activity data 194 on a display.

[0059] ***Description of First Embodiment*** A sales support system 100A will be described with reference to FIG. 10 . The detection model creation unit 112 learns pairs of activity identification data 211A, activity content data 212A, and order record data 220A for each past sales transaction to create a detection model 193. The detection unit 122 receives the activity identification data 231A and activity content data 232A as input and uses the detection model 193 to detect sales activity data 210A of similar transactions. The similar sales activity data 194A is similar business activity data 194 in the sales support system 100A. For example, the similar sales activity data 194A indicates, for each similar sales transaction (a, b), the degree of similarity, transaction name, customer, sales representative, activity date and time, activity type, activity content, and the like. The sales representative refers to the similar sales activity data 194A and performs the next sales activity while taking past similar sales activities into consideration.

[0060] ***Description of Second Embodiment*** The production site support system 100B will be described with reference to FIG. 11 . The detection model creation unit 112 learns pairs of activity identification data 211B and activity content data 212B for each past production plan and creates a detection model 193. The detection unit 122 inputs activity identification data 231B and activity content data 232B for each target production plan and detects production plan data 210B of similar production plans using the detection model 193. The similar production plan data 194B is similar work activity data 194 in the production site support system 100B. For example, the similar production plan data 194B indicates, for each worker (X, Y), the model name, customer, production date and time, planned quantity, worker, activity type, activity content, and the like for each similar production plan (a, b). The worker refers to the similar production plan data 194B and performs production activities while referring to past similar production plans. The activity specification data 211B and activity content data 212B of similar production plans may be displayed by narrowing down the data to the models that are actually allocated according to the results of adjustment by the manager of the production department.

[0061] ***Features of Embodiment 2*** The business support system 100 learns a similar business activity detection model for detecting past business activity data similar to given business activity data. The business support system 100 detects past business activity data similar to a current business activity using the similar business activity detection model.

[0062] Figure 12 shows an overview of the sales support system 100A. The sales support system 100A learns a similar sales activity detection model for detecting past sales activity data similar to given sales activity data. The sales support system 100A detects and presents past sales activity data similar to the current sales activity using the similar sales activity detection model. In Figure 12, the <explanatory variables> are the case name, customer, sales representative, activity type, activity content, etc. The <objective variable> is the sales activity ID. The <output> is the similarity for each sales activity ID. Sales activity information corresponding to the sales activity ID is output by referencing the same reference data as the training data.

[0063] The production site support system 100B learns a similar production plan detection model for detecting past production plan data similar to given production plan data. The production site support system 100B detects and presents past production plan data similar to the current production plan data using the similar production plan detection model.

[0064] ***Effects of the Second Embodiment*** The business support system 100 can present business activity data of similar businesses.

[0065] The sales support system 100A predicts the probability of winning an order based on sales activity data. This makes it possible to predict the probability of winning an order for each sales representative and each customer. By using the activity content data, the accuracy of predicting the probability of winning an order can be improved. The sales department manager can give appropriate instructions to the sales representative based on the probability of winning an order for each business negotiation. The manager can also consider countermeasures for cases with a low probability of winning an order. The sales representative can appropriately select the next sales activity by referring to past sales activity data similar to the input data. By using the activity content data, the accuracy of detecting similar past sales activity data can be improved.

[0066] The production site support system 100B predicts production efficiency based on production plan data. This makes it possible to predict production efficiency for each worker. Then, by using the activity content data for each worker, the accuracy of production efficiency predictions can be improved. The production department manager can improve the efficiency of production operations by adjusting the production plan taking into account differences in production efficiency depending on the combination of worker and machine type. Workers can refer to the activity content included in past similar production plans that are similar to the input data. Then, workers can improve work efficiency by referring to work diaries and on-site improvement activity records related to the work currently assigned to them.

[0067] Third Embodiment A third embodiment of extracting detailed information from activity content data and using the detailed information will be described below, mainly with reference to the differences from the second embodiment, with reference to Figs.

[0068] ***Description of Configuration*** The configuration of the business support system 100 will be described with reference to FIG. 13. The business support system 100 further includes an extraction unit 123. The business support program further causes the computer to function as the extraction unit 123.

[0069] 14 shows the functional configuration of the business support system 100. The extraction unit 123 extracts detailed information based on the content of business activities from the activity content data.

[0070] ***Description of Operation*** The business support method will be described with reference to Fig. 15. In step S310, the extraction unit 123 extracts, for each past business, detailed information based on the content of the business activity from the activity content data 212. The extracted detailed information is referred to as past detailed information.

[0071] For example, if the business activity is a sales activity, the detailed information would be information indicating customer reactions, and if the business activity is a production activity, the detailed information would be information regarding production efficiency.

[0072] The detailed information is extracted using a known method. Examples of known methods include a rule-based method in which a dictionary or rules are created manually, or a method that uses machine learning by preparing a large amount of data and correct labels. Examples of machine learning include CRF and LSTM. CRF is an abbreviation for Conditional Random Field. LSTM is an abbreviation for Long Short Term Memory.

[0073] In step S320, the prediction model creation unit 111 learns the set of the activity identification data 211, the activity content data 212, the task performance data 220, and the past detailed information for each past task, and creates the prediction model 191. Step S320 differs from step S110 in the first embodiment in that the past detailed information is included in the learning data.

[0074] In step S330, the detection model creation unit 112 learns a set of the activity identification data 211, the activity content data 212, and the past detailed information for each past task, and creates the detection model 193. Step S330 differs from step S220 in the second embodiment in that the past detailed information is included in the learning data.

[0075] In step S340, the prediction unit 121 receives the business activity data 230. The business activity data 230 includes activity specification data 231 and activity content data 232. Step S340 is the same as step S120 in the first embodiment.

[0076] In step S350, the extraction unit 123 extracts detailed information based on the content of the business activity from the activity content data 232. The extraction method is the same as the method in step S310. The extracted detailed information is referred to as target detailed information.

[0077] In step S360, the prediction unit 121 receives the business activity data 230 and the target detailed information as input and predicts the evaluation index for the target business using the prediction model 191. Step S360 differs from step S230 in the first embodiment in that the target detailed information is included in the input data.

[0078] In step S370, the prediction unit 121 outputs the evaluation index data 192. Step S370 is the same as step S140 in the first embodiment.

[0079] In step S380, the detection unit 122 receives the business activity data 230 and the target detailed information as input and uses the detection model 193 to detect business activity data 210 of similar business from among the multiple business activity data 210. Step S380 differs from step S260 in the second embodiment in that the target detailed information is included in the input data.

[0080] In step S390, the detection unit 122 outputs the similar work activity data 194. Step S390 is the same as step S270 in the second embodiment.

[0081] ***Description of First Embodiment*** The sales support system 100A will be described with reference to FIG. 16 . The extraction unit 123 extracts customer response information as detailed information from the activity content data 212A for each past sales transaction. The customer response information indicates customer reactions, customer needs, new requests from customers, and the like. The prediction model creation unit 111 learns pairs of the activity identification data 211A, activity content data 212A, order record data 220A, and customer response information for each past sales transaction to create a prediction model 191. The detection model creation unit 112 learns pairs of the activity identification data 211A, activity content data 212A, and customer response information for each past sales transaction to create a detection model 193. The extraction unit 123 extracts customer response information as detailed information from the activity content data 232A. The prediction unit 121 receives the activity identification data 231A, activity content data 232A, and customer response information as input and uses the prediction model 191 to predict the probability of receiving an order for a target project. The detection unit 122 receives the activity identification data 231A, the activity content data 232A, and the customer response information as input, and uses the detection model 193 to detect the sales activity data 210A and the customer response information of similar cases.

[0082] ***Description of Second Embodiment*** A production site support system 100B will be described with reference to FIG. 17 . The extraction unit 123 extracts production efficiency information as detailed information from the activity content data 212B for each past production plan. The production efficiency information is information related to production efficiency, and indicates information obtained by extracting only activities related to the model name of the production plan from the site improvement activity record. The prediction model creation unit 111 learns pairs of activity identification data 211B, activity content data 212B, production performance data 220B, and production efficiency information for each past production plan, and creates a prediction model 191. The detection model creation unit 112 learns pairs of activity identification data 211B, activity content data 212B, and production efficiency information for each past production plan, and creates a detection model 193. The extraction unit 123 extracts production efficiency information as detailed information from the activity content data 232B for each target production plan, and narrows down the activity content data 232B based on the production efficiency information to create narrowed down data. The prediction unit 121 inputs the refined data for each target production plan and predicts the production efficiency of the target production plan for each worker using the prediction model 191. The detection unit 122 inputs the refined data for each target production plan and detects production plan data 210B of similar production plans using the detection model 193.

[0083] ***Features of the Third Embodiment*** The business support system 100 extracts detailed information about the corresponding business from the activity content data attached to the business activity data, and uses this information for learning and inference.

[0084] The sales support system 100A extracts information about customer reactions from the activity content data accompanying the sales activity data, and uses this information for learning and inference.

[0085] The production site support system 100B extracts information about production efficiency from the activity content data accompanying the production plan data, and uses this information for learning and inference.

[0086] ***Effects of the Third Embodiment*** The task support system 100 uses detailed information extracted from the activity content data, thereby improving the accuracy of predicting evaluation indices and the accuracy of detecting similar tasks.

[0087] The sales support system 100A predicts the probability of winning an order based on sales activity data. This makes it possible to predict the probability of winning an order for each sales representative and each customer. By using customer response information in addition to activity content data, the accuracy of predicting the probability of winning an order can be improved. The sales department manager can issue appropriate instructions to the sales representative based on the probability of winning an order for each business negotiation. The manager can also consider countermeasures for cases with a low probability of winning an order. The sales representative can appropriately select the next sales activity by referring to past sales activity data similar to the input data. By using customer response information in addition to activity content data, the accuracy of detecting similar past sales activity data can be improved.

[0088] The production site support system 100B predicts production efficiency based on production plan data. This makes it possible to predict production efficiency for each worker. By using production efficiency information in addition to the activity data of each worker, the accuracy of production efficiency predictions can be improved. The production department manager can improve the efficiency of production operations by adjusting the production plan taking into account differences in production efficiency depending on the combination of worker and machine type. Workers can refer to the activity details included in past similar production plans that are similar to the input data. Workers can then improve their work efficiency by referring to work diaries and on-site improvement activity records related to the work to be assigned this time. By using production efficiency information in addition to the activity data of each worker, the accuracy of detecting similar production plans can be improved.

[0089] Fourth Embodiment The following describes a mode for predicting the tendency of changes in an evaluation index, focusing on differences from the first to third embodiments.

[0090] ***Description of Configuration*** The configuration of the business support system 100 is the same as that in any one of the first to third embodiments.

[0091] ***Explanation of Operation*** The procedure of the business support method is the same as the procedure in any one of the first to third embodiments. However, the operation of the prediction unit 121 differs from the operation in the first to third embodiments.

[0092] The operation of the prediction unit 121 will be described. A period during which business activities for a target business are performed is referred to as a target performance period. The prediction unit 121 divides the target performance period into multiple partial periods. The target performance period includes multiple partial periods. For each partial period, the prediction unit 121 inputs data for the partial period from the activity identification data 231 and data for the partial period from the activity content data 232, and predicts evaluation indicators for the target business activities for the partial periods using a prediction model 191. The prediction unit 121 outputs evaluation indicator data 192. The evaluation indicator data 192 indicates evaluation indicators for the target business activities for each partial period.

[0093] If the business support system 100 includes the extraction unit 123 (see embodiment 3), the extraction unit 123 extracts detailed information for each partial period from the data for that partial period in the activity content data 232. Then, the prediction unit 121 inputs the data for that partial period in the activity identification data 231, the data for that partial period in the activity content data 232, and the detailed information for that partial period, and performs prediction using the prediction model 191 for each partial period.

[0094] ***Features of the Fourth Embodiment*** The business support system 100 divides business activity data into time series, predicts each evaluation index, and visualizes the time series changes in the evaluation index.

[0095] The sales support system 100A divides the current sales activity data into predetermined time series, predicts the order probability for each, and visualizes the time-series change in the order probability.

[0096] The sales support system 100A does not predict the probability of winning an order all at once using the current sales data of a certain negotiation, but predicts the probability of winning by dividing the sales data into a time series of a fixed period (or a variable period). This makes it possible to visualize the upward and downward trends in the probability of winning. For example, only the initial value of the "period" is set manually, and the remaining divisions are performed automatically. For the "period," if the degree of increase or decrease is greater than a threshold, the length of the period may be made variable depending on the degree of increase or decrease in the probability of winning. This makes it possible to shorten the period and increase the resolution of the visualization. For the "period," the period may be gradually shortened from the past to the present. This makes it possible to visualize recent changes in more detail.

[0097] ***Effects of the Fourth Embodiment*** By detecting a business negotiation in which the probability of winning an order is on a downward trend, the manager can quickly consider measures to increase the probability of winning an order and give instructions to the sales representative.

[0098] *** Supplementary Information about the Embodiment *** The hardware configuration of the business support system 100 will be described with reference to Fig. 18. The business support system 100 includes a processing circuit 109. The processing circuit 109 is hardware that realizes a prediction model creation unit 111, a detection model creation unit 112, a prediction unit 121, a detection unit 122, and an extraction unit 123. The processing circuit 109 may be dedicated hardware, or may be a processor 101 that executes a program stored in a memory 102.

[0099] When the processing circuit 109 is dedicated hardware, the processing circuit 109 may be, for example, a single circuit, a multiple circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof. ASIC is an abbreviation for Application Specific Integrated Circuit. FPGA is an abbreviation for Field Programmable Gate Array.

[0100] The business support system 100 may include multiple processing circuits that replace the processing circuit 109.

[0101] In the processing circuit 109, some functions may be realized by dedicated hardware, and the remaining functions may be realized by software or firmware.

[0102] In this way, the functions of the business support system 100 can be realized by hardware, software, firmware, or a combination of these.

[0103] Each embodiment is an example of a preferred embodiment and is not intended to limit the technical scope of the present disclosure. Each embodiment may be implemented in part or in combination with other embodiments. Procedures described using flowcharts, etc. may be modified as appropriate.

[0104] The "part" of each element of the business support system 100 may be read as a "process," a "step," a "circuit," or a "circuitry."

[0105] Aspects of the present disclosure are described below as appendices. (Appendix 1) A business support system including: a prediction unit that receives as input target activity identification data indicating information identifying a target business activity performed in a target business and target activity content data indicating content of the target business activity, and predicts an evaluation index for the target business activity using a prediction model.

[0106] (Supplementary Note 2) The business support system according to Supplementary Note 1, further comprising a prediction model creation unit that learns a set of activity specification data, activity content data, and business performance data for each past business and creates the prediction model.

[0107] (Supplementary Note 3) The business support system according to Supplementary Note 1, further comprising an extraction unit that extracts detailed information based on the content of the target business activity from the target activity content data as target detailed information, and the prediction unit receives the target activity identification data, the target activity content data, and the target detailed information as inputs and predicts the evaluation index for the target business activity using the prediction model.

[0108] (Supplementary Note 4) The business support system according to Supplementary Note 3, further comprising: a prediction model creation unit; wherein the extraction unit extracts detailed information based on the content of business activities from activity content data for each past business as past detailed information; and the prediction model creation unit learns a set of activity identification data, the activity content data, the past detailed information, and business performance data for each past business to create the prediction model.

[0109] (Supplementary Note 5) The business support system according to any one of Supplementary Note 1 to Supplementary Note 4, further comprising a detection unit that receives the target activity identification data and the target activity content data as input and uses a detection model to detect business activity data of a similar business that is similar to the target business from among a plurality of business activity data each including activity identification data and activity content data.

[0110] (Supplementary Note 6) The task support system according to Supplementary Note 5, further comprising a detection model creation unit that learns a set of activity specification data and activity content data for each past task and creates the detection model.

[0111] (Appendix 7) The business support system according to Appendix 5 includes an extraction unit that extracts detailed information based on the content of the target business activity from the target activity content data as target detailed information, and the detection unit receives the target activity identification data, the target activity content data, and the target detailed information as input and uses the detection model to detect the business activity data of the similar business.

[0112] (Appendix 8) The business support system according to Appendix 7 includes a detection model creation unit, wherein the extraction unit extracts detailed information based on the content of business activities from activity content data for each past business as past detailed information, and the detection model creation unit learns a set of activity identification data, the activity content data, and the past detailed information for each past business to create the detection model.

[0113] (Supplementary Note 9) The business support system described in any one of Supplementary Note 1 to Supplementary Note 8, wherein the prediction unit, for each partial period included in a target performance period in which the target business activity was performed, receives as input data for the partial period from the target activity identification data and data for the partial period from the target activity content data, and predicts an evaluation index for the target business activity for the partial period using the prediction model.

[0114] (Appendix 10) The business support system is a system for supporting sales operations, wherein the target activity identification data indicates information identifying a target sales activity carried out in sales of a target case, the target activity content data indicates the content of the target sales activity, and the prediction unit predicts the probability of winning the order for the target case. The business support system described in any one of Appendices 1 to 9.

[0115] (Supplementary Note 11) The business support system described in any one of Supplementary Note 1 to Supplementary Note 9, wherein the business support system is a system for supporting production business, the target activity identification data indicates information that identifies a target production activity carried out in a target production plan, the target activity content data indicates content of the target production activity, and the prediction unit predicts production efficiency of the target production plan.

[0116] (Supplementary Note 12) A business support method that uses target activity identification data indicating information identifying a target business activity carried out in a target business and target activity content data indicating the content of the target business activity as input, and predicts evaluation indicators for the target business activity using a prediction model.

[0117] (Supplementary Note 13) A business support program for causing a computer to execute a prediction process that uses a prediction model to predict evaluation indicators for a target business activity using target activity identification data indicating information that identifies a target business activity carried out in a target business and target activity content data indicating the content of the target business activity as input.

[0118] 100 Business support system, 100A Sales support system, 100B Production site support system, 101 Processor, 102 Memory, 103 Auxiliary storage device, 104 Communication device, 105 Input / output interface, 109 Processing circuit, 111 Prediction model creation unit, 112 Detection model creation unit, 121 Prediction unit, 122 Detection unit, 123 Extraction unit, 190 Storage unit, 191 Prediction model, 192 Evaluation index data, 192A Order probability data, 192B Production efficiency data, 193 Detection model, 194 Similar business activity data, 194A Similar sales activity data, 194B Similar production plan data, 200 Business database, 200A Sales database, 200B Production database, 201 Business data, 201A Sales data, 201B Production data, 210 Business activity data, 210A Sales activity data, 210B Production plan data, 211 Activity specific data, 211A Activity specific data, 211B Activity specific data, 212 Activity content data, 212A Activity content data, 212B Activity content data, 220 Business performance data, 220A Order performance data, 220B Production performance data, 230 Business activity data, 230A Sales activity data, 230B Production plan data, 231 Activity specific data, 231A Activity specific data, 231B Activity specific data, 232 Activity content data, 232A Activity content data, 232B Activity content data.

Claims

1. A business support system comprising a prediction unit that receives as input target activity identification data indicating information identifying target business activities carried out in a target business and target activity content data indicating the content of the target business activities, and predicts evaluation indicators for the target business activities using a prediction model.

2. The business support system according to claim 1, further comprising a prediction model creation unit that learns a set of activity specification data, activity content data, and business performance data for each past business and creates the prediction model.

3. The business support system according to claim 1, further comprising an extraction unit that extracts detailed information based on the content of the target business activity from the target activity content data as target detailed information, and the prediction unit receives the target activity identification data, the target activity content data, and the target detailed information as inputs and predicts the evaluation index for the target business activity using the prediction model.

4. The business support system according to claim 3, further comprising a predictive model creation unit, wherein the extraction unit extracts detailed information based on the content of business activities from activity content data for each past business as past detailed information, and the predictive model creation unit learns a set of activity identification data, the activity content data, the past detailed information, and business performance data for each past business to create the predictive model.

5. A business support system as claimed in any one of claims 1 to 4, comprising a detection unit that uses a detection model with the target activity identification data and the target activity content data as input to detect business activity data of a similar business that is similar to the target business from among a plurality of business activity data each containing activity identification data and activity content data.

6. The task support system according to claim 5, further comprising a detection model creation unit that learns pairs of activity specification data and activity content data for each past task and creates the detection model.

7. The business support system according to claim 5, further comprising an extraction unit that extracts detailed information based on the content of the target business activity from the target activity content data as target detailed information, and the detection unit receives the target activity identification data, the target activity content data, and the target detailed information as inputs and uses the detection model to detect the business activity data of the similar business.

8. The business support system according to claim 7, further comprising a detection model creation unit, wherein the extraction unit extracts detailed information based on the content of business activities from activity content data for each past business as past detailed information, and the detection model creation unit learns a set of activity identification data, the activity content data, and the past detailed information for each past business to create the detection model.

9. A business support system as described in any one of claims 1 to 8, wherein the prediction unit, for each partial period included in the target implementation period in which the target business activity was carried out, inputs data for the partial period from the target activity identification data and data for the partial period from the target activity content data, and predicts an evaluation index for the target business activity for the partial period using the prediction model.

10. The business support system according to any one of claims 1 to 9, wherein the business support system is a system for supporting sales operations, the target activity identification data indicates information identifying the target sales activity carried out in sales of a target case, the target activity content data indicates the content of the target sales activity, and the prediction unit predicts the probability of winning an order for the target case.

11. The business support system according to any one of claims 1 to 9, wherein the business support system is a system for supporting production operations, the target activity identification data indicates information identifying the target production activity carried out in a target production plan, the target activity content data indicates the content of the target production activity, and the prediction unit predicts the production efficiency of the target production plan.

12. A business support method that uses target activity identification data indicating information identifying a target business activity carried out in a target business and target activity content data indicating the content of the target business activity as input, and predicts evaluation indicators for the target business activity using a prediction model.

13. A business support program for causing a computer to execute a prediction process that uses a prediction model to predict evaluation indicators for a target business activity using target activity identification data indicating information that identifies the target business activity carried out in the target business and target activity content data indicating the content of the target business activity as input.