JP7611184B2 - Business support system and business support method - Google Patents

Description

The present invention relates to a business support system and a business support method.

Patent Document 1 describes an office navigation system that is configured to enable an accounting system screen to be called up from an office work terminal on a navigation screen that displays the procedures for office work. The office navigation system includes an office work terminal and an office navigation server that displays on the office work terminal to guide the office work. The office navigation server generates a navigation screen according to an office flow definition that defines the procedures for office work in a financial institution, and when an accounting system screen is used as a navigation screen in the office flow definition, it requests the office work terminal to call up the accounting system screen. The office work terminal stores the accounting system screen, displays the generated navigation screen on a display device, and displays the accounting system screen stored in the memory unit on the display device.

Patent Document 2 describes a sales operations management system designed to support sales activities required at each point in time from inquiry to order acceptance. The sales operations management system manages operations by dividing the order acceptance process from inquiry to order acceptance into a predetermined number of steps for each product type of sales product, creates a file for each case that associates the case identification information of the case input in response to the inquiry with the product information and product type information of the case and customer information for each case, determines a sales representative according to at least the product type of the case based on data in a representative database, updates the contents of the case file, outputs information required for the next meeting with the customer based on information in the meeting preparation content information file, and creates order probability information in the case file that sets achievement/non-achievement information for item information.

JP 2014-127141 A JP 2018-55268 A

When performing work in organizations such as companies and government agencies, analyzing and using historical information on past work (hereinafter referred to as "log information") is effective in improving work efficiency and the success rate of the work. However, in order to obtain useful information and suggestions about the work being considered for how to proceed, it is necessary to identify the variations in the work being considered (hereinafter referred to as "work patterns") and then obtain appropriate information from the log information. In particular, when an organization handles many types and procedures of work and there are a wide variety of work patterns, it is not necessarily easy to identify the work patterns for the work being considered.

The business navigation system described in Patent Document 1 presents navigation screens and accounting system screens to operators in accordance with a business flow definition that defines the procedures for business operations in a financial institution. However, the business navigation system does not identify the patterns of business operations to be considered and provide useful information or suggestions from log information.

In addition, the sales operations management system described in Patent Document 2 determines a sales representative according to the product type of the case, and provides the sales representative with information necessary for the next meeting with the customer and information on the probability of receiving the order. However, the case is specified by a person, and the system does not specify the pattern of the business to be considered, nor does it provide useful information or suggestions from the log information.

The present invention has been made in light of this background, and aims to provide a business support system and a business support method that can obtain and provide useful information about the business being considered from log information.

One aspect of the present invention for achieving the above object is a business support system configured using an information processing device having a processor and a storage device, and storing a business pattern estimation model which is generated based on information associating a business flow, which is information indicating the flow of a series of tasks performed in relation to a business, with a business pattern, which is information indicating a type of business flow corresponding to the series of tasks, and which is a model for identifying the business pattern corresponding to history information, which is information indicating the history of tasks related to the business, and pattern-specific sequence information including information for classifying, for each of the business patterns, the sequences constituting the corresponding business flow according to the influence each of them has on the results of business performance, and target history information which is history information of the business to be supported, and identifying the business pattern corresponding to the target history information using the business pattern estimation model,
Based on the pattern sequence information of the identified task pattern, task execution support information, which is information for supporting the execution of a task, is generated.

Other problems and solutions disclosed in this application will be made clear in the detailed description of the invention and the drawings.

According to the present invention, useful information about the business under review can be obtained from log information and provided.

FIG. 1 is a system flow diagram showing a main configuration of a business support system. 11 is an example of log information. 1 is an example of task pattern log information. 13 is a flowchart illustrating a business flow analysis process. FIG. 13 is a graph showing an example of analysis results (sequence information for each pattern). FIG. 13 is a diagram showing an example of analysis results (sequence information for each pattern) in a table format. 11 is an example of an analysis result (statistical information). FIG. 2 is a system flow diagram showing a configuration related to a business execution support function of the business support system. FIG. 1 is a system flow diagram showing a configuration related to a generation (learning) function of a task pattern estimation model of the task support system. 1 is an example of a learning model used for learning a task pattern estimation model. 13 is an example of a task pattern estimation result. 13 is a flowchart illustrating a history analysis process. 13 is an example of an analysis result (sequence information for each pattern). 13 is an example of an analysis result (sequence information for each pattern). 13 is an example of a support information presentation screen. 1 is a diagram illustrating an example of a hardware configuration of an information processing device used in the configuration of a business support system.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings as appropriate. Note that the following description and drawings are merely examples for explaining the present invention, and are omitted and simplified as appropriate for clarity of explanation. The present invention can be implemented in various other forms. Unless otherwise specified, each component may be singular or plural.

In the following explanation, the same reference numerals may be used for identical or similar configurations, and duplicate explanations may be omitted. In addition, in the following explanation, the letter "S" placed before a reference numeral means a processing step. Furthermore, various information (data) in the following explanation may be expressed or managed in a manner other than the data structure shown as an example. Furthermore, in the following explanation, when explaining various types of identification information, expressions such as "identifier" and "ID" are used as appropriate, but these are interchangeable.

Figure 1 is a system flow diagram showing the general configuration of an information processing system (hereinafter, referred to as "business support system 1") shown as a first embodiment. Business support system 1 is configured using one or more information processing devices (computers).

The business support system 1 acquires information about a series of work flows related to business (information indicating the task owner, work content, work order, work date and time, and the result of the business flow (hereinafter referred to as "business execution result"), etc., hereinafter referred to as "business flow") based on log information, which is text data including information about the history of business performed in the past, classifies the business flow into multiple variations (hereinafter referred to as "business patterns") based on the acquired information, and provides information to support business performed by a user or the organization to which the user belongs by analyzing the relationship between the business flow and the business execution result for each business pattern.

The type of business that the business support system 1 supports is not necessarily limited, but for example, it is business performed by employees, staff, or other task subjects in organizations such as companies, government agencies, and local governments. Note that the task subjects do not necessarily have to be people, and may be robots, etc.

The log information is obtained, for example, from an information processing system operated in an organization such as the one described above. The information processing systems are, for example, a sales management system, a travel expense settlement system, a time attendance management system, an inventory management system, an order management system, a customer management system, a human resources management system, a logistics management system, an accounting system, and a settlement system. In this embodiment, a case where the information processing system is a sales management system will be described as an example.

In the following description, it is assumed that the log information used by the business support system 1 has been pre-processed, such as by formatting and cleansing, using a log parser, log adapter, etc., as appropriate based on the characteristics of each log. Furthermore, it is assumed that the log information has been pre-processed to make a finite series of tasks in the business, from the start task to the completion task to the final task performed in the business, identifiable as a series of business flows based on the actual execution order. Note that business flows can be obtained using publicly known techniques, such as the method described in JP 2017-227944 A, for example.

Each business flow is assigned information indicating the results of the business performed according to the business flow in preprocessing (hereinafter referred to as "business performance result"). The form of the business performance result is not necessarily limited, and may be expressed as a binary value (success/failure) or classified according to a threshold (e.g., quick, delayed, failure, etc.). The business performance result is obtained, for example, based on the content of each task (each task step) included in the business flow or the content of the log information that is the basis of the business flow (work time, etc.). The business performance result may be obtained, for example, by merging the content from another file in which the business performance result is defined.

The business support system 1 analyzes a business flow consisting of a combination of various tasks included in the log information, and automatically classifies the business flow into a business pattern based on the characteristics of the business flow obtained by analyzing the business flow. The business support system 1 classifies the business flow into a business pattern based on, for example, feature quantities (feature values) obtained by quantizing information about the business flow (the results of each task constituting the task, the task subject, the elapsed time, a set of tasks, the order of tasks, etc.). The business support system 1 classifies the business flow into a business pattern according to, for example, information expressed in the specific business flow of each task (information explicitly or implicitly expressed in the business flow, such as the difference in the proficiency of the task subject, the situation of different business flows sharing a common information processing system, whether or not it is a business flow on the customer's side, and information on the difference in the work object (the product or field handled, etc.)).

The business support system 1 classifies the task sequences that make up a business flow according to the impact each sequence has on the results of business execution. Specifically, the business support system 1 classifies business flows into a group (hereinafter referred to as the "successful business flow group") of sequences in which the business is successful (hereinafter referred to as the "successful business flow group") and a group (hereinafter referred to as the "failed business flow group") of sequences in which the business is unsuccessful (hereinafter referred to as the "failed business flow group"). The business support system 1 then identifies frequent sequences that appear only in the successful business flow group (hereinafter referred to as the "successful sequence") and frequent sequences that appear only in the unsuccessful business flow group (hereinafter referred to as the "failed sequence"), as well as frequent sequences that appear in both the successful business flow group and the unsuccessful business flow group (hereinafter referred to as the "common sequence").

As shown in FIG. 1, the business support system 1 includes the functions of a quantization unit 120, a business classification unit 130, and an analysis unit 150. The business support system 1 also includes a memory unit 110 (not shown) that stores log information 111, quantization log information 112, business pattern log information 114, and analysis results 115.

Figure 2 shows an example of log information 111. The illustrated log information 111 is composed of multiple records (entries) having the following items: business flow ID 1111, sequence 1112, task ID 1113, task result 1114, task owner ID 1115, and task date and time 1116. One record of log information 111 corresponds to one business flow.

Of the above items, the business flow ID 1111 stores a business flow ID that is an identifier for the business flow. The order 1112 stores information indicating the order in which each task constituting the business flow is performed (hereinafter referred to as the "task order"). In this example, the task order is a natural number, and tasks are performed in order from the smallest task order. The task ID 1113 stores a task ID that is information identifying each task constituting the business flow. The task result 1114 stores information indicating the result of the task of the task ID (result of each task) (hereinafter referred to as the "task result"). The task subject ID 1115 (worker) stores an identifier (worker name, etc.) of the task subject who performed the task of the task ID (hereinafter referred to as the "task subject ID"). The task date and time 1116 stores the date and time when the task was performed (hereinafter referred to as the "task date and time").

Returning to FIG. 1, the quantization unit 120 generates quantized log information 112, which is information obtained by quantizing information about each business flow obtained from the log information 111. The quantization method is not necessarily limited, but quantization is performed using, for example, BoW (Bag of Words), TF-IDF (Term Frequency - Invers Document Frequency), Word2Vec (CBOW (Continuous Bag of Word), Skip-Gram), etc.

The task classification unit 130 classifies the task flow into a plurality of task patterns based on the quantization log information calculated by the quantization unit 120. The number of task patterns into which the task flow is classified may be preset, or may be received from a user (analyst) who uses the analysis result 115. The number of task patterns is, for example, a number that the user can handle (for example, about 20). The task flow can be classified into task patterns using, for example, K-means or a Gaussian mixture model. When the number of dimensions to be handled is large, dimension reduction may be performed using, for example, latent semantic analysis (LSI (Latent Semantic Indexing), LSA (Latent Semantic Analysis)) or singular value decomposition (SVD (Singular Value Decomposition)).

Figure 3 shows an example of the business pattern log information 114. The illustrated business pattern log information 114 is composed of multiple records (entries) having the following items: business pattern ID 1141, business flow ID 1142, and business execution result 1143.

Of the above items, business pattern ID 1141 stores a business pattern ID that is an identifier for the business pattern. Business flow ID 1142 stores a business flow ID of a business flow that constitutes the business pattern. Each business flow that constitutes a business pattern is one of the features of the business pattern. Business execution result 1143 stores the business execution result of the business flow identified by the business flow ID.

Returning to FIG. 1, the analysis unit 150 performs an analysis for each business pattern in the business pattern log information 114, and generates and outputs an analysis result 115 for each business pattern.

Figure 4 is a flowchart explaining the process for the above analysis performed by the analysis unit 150 (hereinafter referred to as "business flow analysis process S400"). Below, the business flow analysis process S400 will be explained with reference to this figure.

First, the analysis unit 150 selects one business pattern from the business pattern log information 114 (S411).

Next, the analysis unit 150 selects one business flow for the selected business pattern (S412).

The analysis unit 150 then refers to the log information 111 and determines whether the selected workflow is a successful workflow or a failed workflow (S413). If the selected workflow is a successful workflow (S413: successful), the analysis unit 150 stores the selected workflow as a successful workflow (S414). Processing then proceeds to S416. On the other hand, if the selected workflow is a failed workflow (S413: failed), the analysis unit 150 stores the selected workflow as a failed workflow (S415). Processing then proceeds to S416.

In S416, the analysis unit 150 determines whether or not all the business flows of the selected business pattern have been selected. If all the business flows of the selected business pattern have not been selected (S416: NO), the process returns to S412. On the other hand, if all the business flows of the selected business pattern have been selected (S416: YES), the process proceeds to S417.

In S417, the analysis unit 150 identifies frequent sequences for each of the stored successful workflow groups and failed workflow groups using an algorithm for identifying frequent sequences. A frequent sequence here refers to a task and a sequence of tasks that appear frequently in the set. A high frequency is determined by a certain threshold value. For example, if the threshold value is a frequency of "0.8" and a node "initial visit_not so good" appears in the set with a frequency of "0.9", "initial visit_not so good" is considered to be a frequent sequence of length "1". Also, if a node "decide on proposal system" appears after "initial visit_not so good" with a frequency of "0.9", it is considered to be a frequent sequence with an order relationship from "initial visit_not so good" to "decide on proposal system". A frequent sequence represents the degree of frequency in the relative order relationship between nodes. If no upper limit on the distance between nodes is specified, then when comparing with actual cases, the nodes are determined to be frequent if the order relationship holds in a number of cases exceeding a threshold, regardless of how many other nodes are included between the nodes.Frequent sequences can be identified, for example, by using a known method for extracting frequent sequences (such as frequent pattern mining).

Next, the analysis unit 150 identifies the aforementioned "success series," "failure series," and "common series," and outputs the identified results (S418).

In S419, the analysis unit 150 determines whether all business patterns have been selected. If there are unselected business patterns (S419: NO), the process returns to S411. If all business patterns have been selected (S419: YES), the business flow analysis process S400 ends.

Figures 5A and 5B show an example of the analysis results 115 for a certain business pattern. Figure 5A shows the "success sequences," "failure sequences," and "common sequences" (hereinafter also referred to as "pattern-specific sequence information") identified for that business pattern in a graph format (directed graph format), and Figure 5B shows the pattern-specific sequence information in a table format.

In Figure 5A, the circle representing each node lists the task and the results of that task. Note that the results of the task may not always be available. The arrows connecting each node indicate the relative order relationship between the series. The solid arrows represent the "common series", the dashed arrows represent the "success series", and the dashed dotted arrows represent the "failure series".

By having the business support system 1 visually display, for example, the contents of Figures 5A and 5B (the "success sequence," "failure sequence," and "common sequence"), the user can easily obtain information about the business flow that led to success or failure, which can be useful for business improvement plans, etc. For example, the user can easily obtain information about how to proceed with the business after the current task in order to lead to success.

The business support system 1 compares the history information 118 of the tasks performed by the user during the task with the analysis results 115 (pattern-specific sequence information) to identify the task currently being performed by the user, and generates and provides to the user information (hereinafter referred to as "business performance support information") that serves as a guide or reference for deciding which task to perform after the task currently being performed by the user (hereinafter referred to as the "business performance support function"). The business performance support information is information that indicates, for example, tasks that should be performed in the future (hereinafter referred to as "recommended tasks"), tasks that should be performed early (hereinafter referred to as "early tasks"), tasks that should be performed after the right timing (hereinafter referred to as "premature tasks"), tasks that should be avoided (hereinafter referred to as "avoided tasks"), tasks that should be observed (not actively pursued) (hereinafter referred to as "observation tasks").

Returning to FIG. 1, the analysis unit 150 further performs statistical processing for each business pattern in the business pattern log information 114 to generate and output analysis results 115 for each business pattern.

Figure 5C shows an example of the analysis result 115 obtained by the above statistical processing. Each record (entry) of the illustrated analysis result 115 corresponds to one business pattern. The analysis result 115 includes multiple entries (records) having the following items: business pattern ID 1151, number of executions 1152, order odds 1153, order received/lost 1154, and classification result 1155. Note that the number of executions 1152, order odds 1153, and order received/lost 1154 store the results of statistical processing of the work results 1114 in the log information 111.

A business pattern ID is stored in the business pattern ID 1151. The number of executions 1152 stores the number of business flows belonging to the business pattern. The order odds 1153 stores the value of odds calculated for the business pattern using a predetermined formula. The number of orders (successful) and the number of orders (failed) are stored in the order received/lost 1154. The classification result 1155 stores information indicating the classification of the business pattern (promising group, possible group, hopeless group) according to the value of the order odds 1153. In this example, the value (first threshold) of the order odds 1153 that distinguishes between the promising group and the possible group is set to "3.0", and the value (second threshold) of the order odds 1153 that distinguishes between the possible group and the hopeless group is set to "4.0".
The value of 53 (second threshold) is set to “0.10”.

By being provided with the analysis results 115 shown in the figure, the user (analyst) can obtain, for example, suggestions as to which business pattern of business flow the work should be followed. For example, the user can effectively improve the restructuring of business processes and the method of instructing workers, etc.

Figure 6 is a system flow diagram explaining the configuration related to the above-mentioned business execution support function of the business support system 1. The business support system 1 further includes a business pattern estimation unit 160 and a history analysis processing unit 170. The storage unit 110 further stores a business pattern estimation model 155, which is a machine learning model, history information 118, and business execution support information 119.

History information 118 includes information about the work that the user has performed up to the present in the course of business. The method of acquiring history information 118 is not necessarily limited, but it may be acquired, for example, by manual input by the user or from various information processing systems used by the user.

The task pattern estimation unit 160 inputs the history information 118 to the task pattern estimation model 155 to identify a task pattern corresponding to the history information 118 and generate a task pattern estimation result 117 including information indicating the identified task pattern. The task pattern estimation model 155 is, for example, a deep neural network (DNN), a long short-term memory (LSTM), or the like.
) and the like. The generation (learning) of the task pattern estimation model 155 will be described later.

The history analysis processing unit 170 generates business execution support information 119 based on the analysis results 115 corresponding to the business patterns described in the business pattern estimation results 117.

Figure 7 is a system flow diagram explaining the main configuration related to the generation (learning) function of the business support system 1 for the business pattern estimation model 155. The business support system 1 further includes a learning data generation unit 165 and a learning processing unit 166. The storage unit 110 further stores a label 173 and learning data 175.

The learning data generation unit 165 generates learning data 175 based on the log information 111 and the label 173. The label is set by the user via a user interface, for example, but may be set based on the business pattern log information 114, for example. The learning data 175 is information that associates all or part of each business flow in the log information 111 with a business pattern (business pattern ID) (to which the all or part corresponds).

For example, in the case of an operation with a workflow ID of "GF0001" in Fig. 2, the learning data 175 is information associating the entire workflow (all records with the workflow IDs "initial visit" to "order") with a workflow pattern (work pattern ID), or information associating a part of the workflow (for example, a set of two records with the workflow IDs "initial visit" and "requirements acquisition") with a workflow pattern (work pattern ID). Note that a set of records for a part of a workflow is a record of the operation performed first in the workflow, and a set of records of one or more operations performed sequentially after the operation (in the example of Fig. 2, a set of records including a record with the workflow ID 1113 of "initial visit" and the subsequent records).

Figure 8 shows an example of learning data 175. The example learning data 175 is composed of multiple records having the following fields: learning data ID 1751, business flow ID 1752, sequence 1753, task ID 1754, task result 1755, and business pattern ID 1756.

Of the above items, learning data ID 1751 stores a learning data ID, which is an identifier assigned to each piece of learning data. A group of records with a common learning data ID constitutes one piece of learning data. Business flow ID 1752, sequence 1753, task ID 1754, and task result 1755 are all information obtained from log information 111. Business pattern ID 1756 stores the business pattern ID of the label (business pattern) set for the learning data.

Figure 9 shows an example of a business pattern estimation result 117 obtained by the business pattern estimation unit 160 estimating the business pattern of the input history information 118 (target history information) using the business pattern estimation model 155. As shown in the figure, the business pattern estimation result 117 is composed of one or more records having the items of business pattern ID 1771 and likelihood 1772.

Of the above items, the business pattern ID described above is stored in business pattern ID 1771. The likelihood of the business pattern is stored in likelihood 1772. The likelihood is a value calculated by business pattern estimation model 155 that represents the similarity (similarity) between history information 118 and the business pattern ID, and a larger value indicates a higher similarity.

Figure 10 is a flowchart explaining the process (hereinafter referred to as "history analysis process S1000") in which the history analysis processing unit 170 generates business execution support information 119 based on the business pattern estimation result 117 and the analysis result 115 (pattern-specific sequence information). The history analysis process S1000 will be explained below with reference to the same figure. Note that the following explanation will be given taking as an example a case in which the contents shown in Figures 11A and 11B are acquired as the analysis result 115.

First, the history analysis processing unit 170 selects one business pattern from the business pattern estimation result 117 (S1011). For example, the history analysis processing unit 170 selects the business pattern with the highest likelihood 1172. Note that the history analysis processing unit 170 may select multiple business patterns and perform the following processes S1012 to S1016 for each selected business pattern.

Next, the history analysis processing unit 170 obtains the business flow of the selected business pattern (business flow based on the sequence information for each pattern) from the analysis result 115 (S1012).

Next, the history analysis processing unit 170 compares the history information 118 with the acquired business flow to identify the user's current work in the business flow (S1013). Note that the work sequence in the history information 118 may not completely match the work sequence in the business flow. In that case, for example, the user's current work is identified by sequentially tracing the work that matches the business flow along the sequence.

Next, the history analysis processing unit 170 identifies the work that follows the identified current work (hereinafter referred to as the "successor work") from the business flow obtained in S1012 (S1014).

For example, in the example of FIG. 11A, when the user's current task is identified as "review of proposal policy" in S1013, the history analysis processing unit 170 identifies "estimate", "create proposal", "deal with issue_success", "determine responsibility system", and "acquire requirements_success" as the subsequent tasks. Note that "create proposal", "deal with issue_success", and "determine responsibility system" are all tasks that follow "determine proposal system" like "review of proposal policy" and are tasks that are performed in parallel with "review of proposal policy", so the history analysis processing unit 170 also identifies these tasks as the subsequent tasks of "review of proposal policy". In addition, since there is no task preceding "requirements acquisition_success", the history analysis processing unit 170 also identifies the task as the subsequent task of "review of proposal policy".

In the example of FIG. 11A, if the user's current task identified in S1113 is "Decide on proposal structure," the history analysis processing unit 170 identifies "Create proposal," "Respond to issue - successful," "Decide on responsibility structure," "Consider proposal policy," and "Acquire requirements - successful" as the subsequent tasks.

In the example of FIG. 11A, if the current task of the user identified in S1113 is "Initial visit not so good," the history analysis processing unit 170 identifies "Determine proposal structure" and "Requirements acquisition_success" as subsequent tasks.

Returning to FIG. 10, the history analysis processing unit 170 then generates business execution support information 119 based on the identified subsequent tasks (S1015).

For example, if the transition from the user's current task to the subsequent task is a "success sequence," the history analysis processor 170 generates task execution support information 119 that designates the subsequent task as a "recommended task." For example, in the example of FIG. 11A, if the current state of the user's task is "reviewing the proposal policy," the history analysis processor 170 generates task execution support information 119 that designates the subsequent tasks in the "success sequence," "writing a proposal" and "addressing the issue_success," as "recommended tasks."

For example, if the transition from the user's current task to the subsequent task is a "failure series," the history analysis processing unit 170 generates task execution support information 119 with the content designating the subsequent task as an "avoidable task." For example, in the example of FIG. 11A, if the current state of the user's task is "reviewing the proposed policy," the history analysis processing unit 170 generates task execution support information 119 with the content designating the subsequent task in the failure series, "determining the responsibility system," as an "avoidable task."

For example, for a subsequent task in which the transition from the user's current task to the subsequent task does not form a "common series" or a "failed series," the history analysis processor 170 generates task execution support information 119 with the content being an "early execution task." For example, in the example of FIG. 11A, if the current state of the user's task is "Determine proposed structure," the history analysis processor 170 generates task execution support information 119 with the content being an "early execution task" for the subsequent task "Requirements acquisition_success."

In addition, for example, if there is a possibility that the execution order of the "successful sequence" will be disrupted by executing a specified subsequent task, the history analysis processing unit 170 generates task execution support information 119 with the content that designates the subsequent task as a "premature task." For example, in the example of FIG. 11A, when the current state of the user's task is "Consider proposal policy," there are other subsequent tasks in the "successful sequence" for "Estimate," which is a subsequent task in the common sequence, such as "Write proposal" and "Respond to issue_success," so the history analysis processing unit 170 generates task execution support information 119 with the content that designates the subsequent task "Estimate" as a "premature task."

Also, for example, depending on the business flow, the same subsequent task may be both a "success sequence" and a "failure sequence." In that case, the history analysis processing unit 170 may classify the subsequent task as "
The task execution support information 119 is generated, which indicates that the task is to be performed in a silent manner (not to be actively performed).

In addition, for example, if a subsequent task has a transition to a "failure series", the history analysis processing unit 170 generates task execution support information 119 with content that recommends a subsequent task in a series other than the series of the subsequent task. For example, in the example of FIG. 11A, if the current state of the user's task is "Determine proposed system", the subsequent task "Consider proposed policy" has a transition to a failure series (transition to "Determine responsibility system"), so the history analysis processing unit 170 generates task execution support information 119 with content that recommends the execution of a subsequent task in a series other than the series of the subsequent task (for example, "Determine responsibility system").

Returning to FIG. 10, the history analysis processing unit 170 then provides the user with information based on the generated business execution support information 119 (S1016).

Figure 12 is an example of a screen (hereinafter referred to as the "support information presentation screen 1200") that the history analysis processing unit 170 presents to the user via the user interface when providing the contents of the business execution support information 119 to the user.

As shown in the figure, the support information presentation screen 1200 has a display field 1211 for the person in charge of the work, a display field 1212 for the work pattern, a display field 1213 for the likelihood, a display field 1214 for the current work, a display field 1220 for the work flow, and a display field 1230 for recommendations.

The business person display field 1211 displays the user identifier (business person name) acquired together with the history information 118. The business pattern display field 1212 displays the identifier (business pattern ID) of the business pattern selected in S1011 of FIG. 10. The likelihood display field 1213 displays the likelihood of the business pattern selected in S1011 of FIG. 10 (likelihood 1772 of the business pattern estimation result 177). The current work display field 1214 displays information indicating the current work of the user identified in S1013 of FIG. 10. The business flow display field 1220 displays the business flow acquired in S1012 of FIG. 10. The recommendation display field 1230 displays the business execution support information 119 generated in S1015 of FIG. 10.

By referring to the support information presentation screen 1200, the user can easily check to which task pattern the task the user is currently performing belongs. The user can also check the history of the tasks that he or she has performed and obtain information such as whether the task that the user is currently performing is close to success or failure.

In addition, users can get suggestions on what tasks they should and should not do to complete their work successfully, which can be used as useful information when deciding how to proceed with their work.

For example, the support information presentation screen 1200 may further display information based on the analysis results 115 (statistical information) shown in FIG. 5C (or may be provided with a function for displaying information upon receiving a user instruction). By referring to the above information, the user can, for example, check the success probability (order odds) of the task he or she is currently performing, and appropriately select the task that is likely to lead to the success of the task, and proceed with the task.

Furthermore, when the value of the likelihood 1772 of the task pattern estimation result 177 shown in Fig. 9 is small, there is a high possibility that erroneous information will be provided to the user. Therefore, for example, the task support system 1 may provide the task execution support information 119 to the user when the value of the likelihood 1772 is equal to or greater than a preset threshold value (hereinafter referred to as the "likelihood threshold value"), and may not provide the task execution support information 119 to the user when the value of the likelihood 1772 is smaller than the likelihood threshold value (or may present information indicating that the information is low in reliability when presented to the user).

<Example of information processing device>
FIG. 13 shows an example of the configuration of an information processing device constituting the business support system 1. The illustrated information processing device 10 includes a processor 11, a main memory device 12, an auxiliary memory device 13, an input device 14, an output device 15, and a communication device 16. The illustrated information processing device 10 may be realized, in whole or in part, by using virtual information processing resources provided by a virtualization technology or a process space separation technology, such as a virtual server provided by a cloud system. In addition, all or in part of the functions provided by the information processing device 10 may be realized by a service provided by a cloud system via an API (Application Program Interface), for example. The business support system 1 may be configured by using a plurality of information processing devices 10 connected to each other so as to be able to communicate with each other.

In the figure, the processor 11 is, for example, a CPU (Central Processing Unit),
PU (Micro Processing Unit), GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), ASIC (Application Specific Integrated Circuit)
), AI (Artificial Intelligence) chips, etc.

The main memory device 12 is a device for storing programs and data, and is, for example, a ROM (Read Only Memory).
Only Memory), RAM (Random Access Memory), NVRAM (Non Volatile RAM), etc.

The auxiliary storage device 13 is, for example, a solid state drive (SSD), a hard disk drive, an optical storage device (CD (Compact Disc), DVD (Digital Versatile Disc), etc.), a storage system, an IC card, a read/write device for a recording medium such as an SD card or an optical recording medium, a storage area of a cloud server, etc. Programs and data can be read into the auxiliary storage device 13 via a recording medium reader or a communication device 16. The programs and data stored (memorized) in the auxiliary storage device 13 are read into the main storage device 12 as needed.

The input device 14 is an interface that accepts input from the outside, and is, for example, a keyboard, a mouse, a touch panel, a card reader, a pen-input tablet, a voice input device, etc.

The output device 15 is an interface that outputs various information such as the progress of processing and the results of processing. The output device 15 is, for example, a display device (liquid crystal monitor, LCD (Liquid Crystal Display), graphic card, etc.) that visualizes the various information described above, a device that converts the various information described above into audio (audio output device (speaker, etc.)), or a device that converts the various information described above into text (printer, etc.). Note that, for example, the information processing device 10 may be configured to input and output information to and from other devices via the communication device 16.

The input device 14 and output device 15 constitute a user interface that accepts information from the user and presents information.

The communication device 16 is a device that realizes communication with other devices. The communication device 16 is a wired or wireless communication interface that realizes communication with other devices via a communication medium such as a communication network, and is, for example, a NIC (Network Interface Card), a wireless communication module, a USB module, etc.

The information processing device 10 includes, for example, an operating system, a file system, a database,
A DataBase Management System (MS) (relational database, NoSQL, etc.), a Key-Value Store (KVS), etc. may also be introduced.

The functions of the business support system 1 are realized by the processor 11 reading and executing a program stored in the main memory device 12, or by the hardware (FPGA, ASIC, AI chip, etc.) that constitutes the business support system 1. The business support system 1 stores the various types of information (data) described above, for example, as tables in a database or files managed by a file system.

Although one embodiment of the present invention has been described above, it goes without saying that the present invention is not limited to the above embodiment and can be modified in various ways without departing from the spirit of the invention. For example, the above embodiment has been described in detail to clearly explain the present invention, and is not necessarily limited to having all of the configurations described. In addition, it is possible to add, delete, or replace part of the configuration of the above embodiment with other configurations.

For example, a probabilistic model-based representation such as a Bayesian network may be used to represent a business flow (sequence of tasks).

In addition, the above-mentioned configurations, functional units, processing units, processing means, etc. may be realized in part or in whole by hardware, for example, by designing them as integrated circuits. In addition, the above-mentioned configurations, functions, etc. may be realized in software by a processor interpreting and executing a program that realizes each function. Information such as the program, table, file, etc. that realizes each function can be stored in a memory, a hard disk, a recording device such as an SSD (Solid State Drive), an I/O device, etc.
It can be placed on a recording medium such as a C card, an SD card, or a DVD.

Furthermore, the layout of the various functional units, various processing units, and various databases of each information processing device described above is merely an example. The layout of the various functional units, various processing units, and various databases can be changed to an optimal layout in terms of the performance, processing efficiency, communication efficiency, etc. of the hardware and software equipped in these devices.

In addition, the configuration (schema, etc.) of the database that stores the various types of data mentioned above can be flexibly changed from the perspective of efficient use of resources, improved processing efficiency, improved access efficiency, improved search efficiency, etc.

1 Business support system, 111 Log information, 112 Quantized log information, 114 Business pattern log information, 115 Analysis result, 117 Business pattern estimation result, 118 History information, 119 Business execution support information, 120 Quantization unit, 130 Business classification unit, 150 Analysis unit, 155 Business pattern estimation model, 160 Business pattern estimation unit, 165 Learning data generation unit, 166 Learning processing unit, 170 History analysis processing unit, 173 Label, 175 Learning data, S400 Business flow analysis processing, S1000 History analysis processing, 1200 Support information presentation screen

Claims (15)

The information processing device includes a processor and a storage device.
a business pattern estimation model that is generated based on information that associates a business flow, which is information indicating the flow of a series of tasks performed in relation to a business, with a business pattern, which is information indicating the type of business flow corresponding to the series of tasks, and that is a model that identifies the business pattern that corresponds to history information, which is information indicating the history of tasks related to the business;
pattern sequence information including information in which sequences constituting the corresponding workflows for each of the business patterns are classified according to the influence each of the sequences has on the results of business execution;
Target history information, which is history information of the business to be supported;
Remember,
Identifying the task pattern corresponding to the target history information by the task pattern estimation model;
generating task execution support information, which is information for supporting task execution, based on the pattern sequence information of the identified task pattern;
Business support system. The business support system according to claim 1,
The pattern sequence information includes information indicating at least one of a success sequence, which is a sequence of tasks that leads to success in a business, a failure sequence, which is a sequence of tasks that leads to failure in a business, and a common sequence, which is a sequence that appears in both the success sequence and the failure sequence.
Business support system. The business support system according to claim 2,
Identifying a current task in the business to be supported by comparing the target history information with a business flow based on the pattern sequence information of the business pattern identified for the target history information;
Identifying whether a subsequent task that is a task subsequent to the identified current task belongs to the success sequence, the failure sequence, or the common sequence;
generating the business execution support information having contents corresponding to the identified series;
Business support system. The business support system according to claim 3,
storing log information including the workflow and a workflow execution result, which is information showing a result of executing a workflow according to the workflow;
generating task pattern log information that classifies the task flow into a plurality of task patterns based on the feature amount of the task flow in the log information;
generating the pattern-by-pattern sequence information based on the task pattern log information and the task performance result acquired from the log information;
Business support system. The business support system according to claim 4,
Based on the task pattern log information and the task execution results obtained from the log information, for each of the task patterns, identify whether the task flow belonging to each of the task patterns is a successful task flow that leads to success or a failed task flow that leads to failure;
a frequent sequence that appears only in a successful workflow group that is a group of the successful workflows is identified as the successful sequence, a frequent sequence that appears only in a failed workflow group that is a group of the failed workflows is identified as the failure sequence, and a frequent sequence that appears in both the successful workflow group and the failed workflow group is identified as the common sequence, thereby generating the sequence information for each pattern.
Business support system. The business support system according to claim 3,
generating the task execution support information indicating that the subsequent task should be performed if the subsequent task is a task in the successful sequence;
Business support system. The business support system according to claim 3,
generating the task execution support information indicating that the subsequent task should be avoided if the subsequent task is a task of the failed series;
Business support system. The business support system according to claim 3,
generating the task execution support information indicating that the execution of the subsequent task should be waited and watched if the subsequent task is a task of the common series;
Business support system. The business support system according to claim 1,
the task pattern estimation model is a machine learning model trained using training data including information associating log information including the task flow with a task pattern corresponding to the task flow acquired from the log information;
Business support system. The business support system according to claim 9,
The task pattern estimation model generates a likelihood of the task pattern identified as the target history information when identifying the task pattern corresponding to the target history information;
Identifying the business pattern whose likelihood is equal to or greater than a preset threshold as the business pattern corresponding to the target history information.
Business support system. The business support system according to claim 1,
A user interface is provided.
providing the task execution support information to a user via the user interface;
Business support system. An information processing device having a processor and a storage device,
a business pattern estimation model that is generated based on information that associates a business flow, which is information indicating the flow of a series of tasks performed in relation to a business, with a business pattern, which is information indicating the type of business flow corresponding to the series of tasks, and that is a model that identifies the business pattern that corresponds to history information, which is information indicating the history of tasks related to the business;
pattern sequence information including information in which sequences constituting the corresponding workflows for each of the business patterns are classified according to the influence each of the sequences has on the results of business execution;
Target history information, which is history information of the business to be supported;
storing the
Identifying the task pattern corresponding to the target history information by the task pattern estimation model; and
generating task execution support information, which is information for supporting task execution, based on the pattern sequence information of the identified task pattern;
A business support method for carrying out the above. The business support method according to claim 12,
The pattern sequence information includes information indicating at least one of a success sequence, which is a sequence of tasks that leads to success in a business, a failure sequence, which is a sequence of tasks that leads to failure in a business, and a common sequence, which is a sequence that appears in both the success sequence and the failure sequence.
Business support methods. The business support method according to claim 13,
The information processing device,
A step of identifying a current task in the business to be supported by comparing the target history information with a business flow based on the pattern sequence information of the business pattern identified for the target history information;
A step of identifying whether a subsequent task that is a task subsequent to the identified current task belongs to the success sequence, the failure sequence, or the common sequence; and
generating the task execution support information having contents corresponding to the identified series;
A business support method that further implements the above. The business support method according to claim 14,
The information processing device,
storing log information including the workflow and a workflow execution result which is information indicating a result of executing a workflow according to the workflow;
generating task pattern log information that classifies the task flow into a plurality of task patterns based on the feature amount of the task flow in the log information; and
generating the pattern-by-pattern sequence information based on the task pattern log information and the task performance result acquired from the log information;
A business support method that further implements the above.

Images