JP7758698B2 - Anomaly detection system, anomaly detection method, and anomaly detection program - Google Patents

Description

The present invention relates to an anomaly detection system, an anomaly detection method, and an anomaly detection program.

For example, the construction industry has traditionally been one in which anomalies frequently occur in construction-related data. Because anomalies must be identified from vast amounts of data, the scope of what can be detected manually exceeds that which can be achieved by humans. For example, Patent Document 1 discloses a system for detecting anomalies in business data.

JP 2015-95122 A

However, Patent Document 1 does not mention anything about automatically detecting anomalies in case (business) data and taking early action to identify and deal with them.

In view of the above, the present invention aims to provide an anomaly detection system, an anomaly detection method, and an anomaly detection program that can automatically detect anomalies in case data and enable early discovery and response.

To solve the above-mentioned problems and achieve the objectives, the present invention provides an anomaly detection system equipped with a control unit, which is configured to be able to access case data including one or more specific conditions, progress rates, and case progress rates. It is characterized by comprising: detection means for dividing a population of past case data according to specified specific conditions, calculating an approximate curve for the progress rate and case progress rate of the case data for each population, calculating a predicted value for the case progress rate by substituting the progress rate into the approximate curve function, calculating upper and lower limits based on the approximate curve and the predicted value for the case progress rate, and detecting anomalies by comparing the case data to be detected with the upper and lower limits; and display control means for displaying analysis data on an analysis screen for case data in which an anomaly has been detected.

Furthermore, according to one aspect of the present invention, the project data includes construction data, the specific conditions include the type of work, construction period, or public/private classification, the progress rate includes the construction period progress rate calculated by the number of months elapsed since construction divided by the planned construction period, and the project progress rate includes the cumulative progress rate calculated by the currently incurred costs divided by (budgetary cost + currently incurred costs), and the construction data may include the work, organization, one or more specific conditions, fiscal year and month, sales amount, construction period progress rate, and cumulative progress rate.

Furthermore, according to one aspect of the present invention, the display control means may display a message in a predetermined area of the analysis screen based on the construction data, including the construction work detected as abnormal, the detection method, and the cumulative progress rate at which the abnormality was determined.

Furthermore, according to one aspect of the present invention, the display control means may display, in a predetermined area of the analysis screen based on the construction data, the actual cumulative progress rate of past construction projects by population or of all past construction projects, and its approximate curve, as well as a graph showing the trend in the actual cumulative progress rate associated with construction projects in which an abnormality has been detected, by construction period progress rate.

Furthermore, according to one aspect of the present invention, the display control means may display, in a predetermined area of the analysis screen, the actual cumulative progress rate of past construction work and its approximate curve, and a graph showing the trend in the actual cumulative progress rate associated with construction work in which an abnormality has been detected, by accounting year and month, based on the construction work data.

Furthermore, according to one aspect of the present invention, the display control means may display a table showing actual sales and sales forecast trends by organization for a specified accounting period in a specified area of the analysis screen based on the construction data.

Furthermore, according to one aspect of the present invention, the display control means may display, in a predetermined area of the analysis screen, a table showing the trends in actual sales and sales forecasts for each construction project of the organization in charge of the construction project in which the abnormality occurred, based on the construction project data.

Furthermore, to solve the above-mentioned problems and achieve the objectives, the present invention provides an anomaly detection method executed by an information processing device equipped with a control unit, wherein the control unit is configured to be able to access case data including one or more specific conditions, progress rates, and case progress rates, and the method includes a detection process executed by the control unit to divide a population of past case data according to specified specific conditions, calculate an approximate curve for the progress rate of business data and the case progress rate for each population, calculate a predicted value of the case progress rate by substituting the progress rate into the approximate curve function, calculate upper and lower limits based on the approximate curve and the predicted value of the case progress rate, and compare the case data to be detected with the upper and lower limits to detect anomalies; and a display control process to display analysis data on an analysis screen for case data in which an anomaly has been detected.

Furthermore, to solve the above-mentioned problems and achieve the objectives, the present invention provides an anomaly detection program to be executed by an information processing device having a control unit, wherein the control unit is configured to be able to access case data including one or more specific conditions, progress rates, and case progress rates, and the control unit executes the following detection steps: a detection step in which the control unit divides a population of past case data according to the specified specific conditions, calculates an approximate curve for the progress rate and case progress rate of the case data for each population, calculates a predicted value for the case progress rate by substituting the progress rate into the approximate curve function, calculates upper and lower limit values based on the approximate curve and the predicted value for the case progress rate, and compares the case data to be detected with the upper and lower limit values to detect anomalies; and a display control step in which, for case data in which an anomaly has been detected, analysis data is displayed on an analysis screen.

The present invention has the effect of automatically detecting anomalies in case data, enabling early detection and response.

FIG. 1 is a diagram showing an example of an image of a fraudulent system. FIG. 2 is a diagram showing an example of an image of a good performance system. FIG. 3 is a diagram for explaining the applicable range (industries in which the present invention can be used) of the present invention. FIG. 4 is a diagram for explaining the measures and effects for problem (1). FIG. 5 is a diagram for explaining the measures and effects for problem (2). FIG. 6 is a diagram for explaining the measures and effects for the problem (3). FIG. 7 is a diagram for explaining the measures and effects for the problem (4). FIG. 8 is a diagram showing an example of the display of the initial analysis screen. FIG. 9 is a diagram showing an example of the display of the analysis screen. FIG. 10 is a block diagram illustrating a hardware configuration of the anomaly detection system according to the embodiment. FIG. 11 is a flowchart for explaining an outline of the overall processing of the control unit of the anomaly detection system of this embodiment. FIG. 12 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to this embodiment. FIG. 13 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to this embodiment. FIG. 14 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to this embodiment. FIG. 15 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to this embodiment. FIG. 16 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to this embodiment. FIG. 17 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to this embodiment. FIG. 18A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system according to the present embodiment. FIG. 18B is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 19A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system according to the present embodiment. FIG. 19B is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 20A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system according to the present embodiment. FIG. 20B is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 21 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to this embodiment. FIG. 22A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system according to the present embodiment. FIG. 22B is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 23A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system according to the present embodiment. FIG. 23B is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 24A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system according to the present embodiment. FIG. 24B is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 25A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system according to the present embodiment. FIG. 25B is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 26A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system according to the present embodiment. FIG. 26B is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 27A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system according to the present embodiment. FIG. 27B is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 28A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system according to the present embodiment. FIG. 28B is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 29A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system according to the present embodiment. FIG. 29B is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 30 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 31 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 32 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 33A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system according to the present embodiment. FIG. 33B is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system according to the present embodiment. FIG. 34 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 35 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 36 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 37 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to this embodiment. FIG. 38 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system according to the present embodiment. FIG. 39 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system in this embodiment. FIG. 40 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system in this embodiment. FIG. 41 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system in this embodiment. FIG. 42A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 42B is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 43 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system in this embodiment. FIG. 44 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system in this embodiment. FIG. 45 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 46A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 46B is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 47A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 47B is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 48A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 48B is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 49A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 49B is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 50 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system in this embodiment. FIG. 51 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system in this embodiment. FIG. 52 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 53A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 53B is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 54 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system in this embodiment. FIG. 55 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 56 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 57 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 58A is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 58B is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 59 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system in this embodiment. FIG. 60 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system in this embodiment. FIG. 61 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 62 is a diagram for explaining a specific example of the processing of the control unit of the anomaly detection system in this embodiment. FIG. 63 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 64 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 65 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 66 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 67 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 68 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 69 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 70 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 71 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 72 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 73 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 74 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment. FIG. 75 is a diagram for explaining a specific example of processing by the control unit of the anomaly detection system in this embodiment.

Below, an anomaly detection system according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

[1. Overview]
The outline of the present invention will be explained in the following order: (1-1. Background and premise), (1-2. Measures and effects for solving the problems), and (1-3. Analysis screen).

(1-1. Background/premises)
The background and premise will be explained with reference to Figures 1 to 3. The construction industry has traditionally been an industry where anomalies have frequently occurred in construction-related data. Since it is necessary to check for anomalies from a huge amount of data, the range of what can be checked by human beings is becoming more than what can be checked by human beings, and there is a demand for a system that can detect and deal with anomalies early by means other than human beings.

In this embodiment, a system is constructed that automatically detects abnormal patterns on a regular basis by creating scenarios to detect and deal with data anomalies early and without human intervention.

Specifically, in this embodiment, fraudulent business data is automatically detected and notified within transaction data in the business system. In doing so, the approximate progress of construction work managed by the company is calculated based on past construction results, and construction work that shows a large discrepancy compared to the results of uncompleted construction work is detected as an anomaly. A screen is also output that allows users to check the approximate progress curve of construction work and the progress curve of construction work for which an anomaly has been detected. Additionally, because construction anomalies affect the organization's figures, a screen is output that allows users to check the results by organization and by construction work period.

This embodiment makes it possible to regularly detect fraud from vast amounts of construction-related information, and to check the predicted future progress of unfinished construction work and the degree of workload that may occur in the future. It also makes it possible to perform detailed analysis of project completion forecasts for each organization by period and trends in progress by month for construction work.

As an example, we will assume that this will be used by companies in the construction industry that use "percentage-of-completion sales" to manage the progress of each project.

There are two methods for recording sales for construction work in the construction industry:
(1) Percentage-of-Progress Sales is a method of recording sales in accordance with the ratio of the accumulated cost amount recorded during construction work to the estimated final cost amount.
(2) Sales from completed construction is a method of recording sales when construction is completed.

Possible patterns of data anomalies include the following:
- Possible fraudulent activities include misallocation of construction costs, non-recording of costs, and delays in recording sales.
- For projects with good results, construction may be brought forward.

A feature that applies to both patterns is the "impact on the construction progress rate." The approximation line of the progress rate trends of past construction work is used as a reference value, representing "the progress rate trends that our company basically follows when carrying out construction work." If there is a large discrepancy between this approximation line of the progress rate trends of past construction work and the current construction work, this can be seen as an abnormality, as it indicates that the progress rate is different from past trends.

Figure 1 is a diagram showing an example of fraudulent activity. Figure 1 shows a graph of the progress curve, with the horizontal axis representing the progress rate (%) for the construction period and the vertical axis representing the cumulative progress rate (%). An approximation curve of the progress rate trends for past construction work is plotted alongside the progress rate of the work. In this diagram, the progress rate is abnormally low compared to the approximation curve, which may indicate fraudulent activity.

Figure 2 is a diagram showing an example of what a good performance type might look like. Figure 2 shows a graph of the progress curve, with the horizontal axis representing the progress rate (%) for the construction period and the vertical axis representing the cumulative progress rate (%). An approximation curve of the progress rate trends for past construction work is plotted along with the progress rate of the construction work. In this figure, the progress rate for the construction work has increased abnormally compared to the approximation curve, which may indicate a good performance type.

Here, the progress rate of the construction period, the completed volume, and the progress rate of the construction work (cumulative progress rate) will be explained.
・The progress rate for the construction period refers to the progress rate for the construction period in months as of the current accounting month. For example, if there is a construction project with a 12-month construction period, which begins in January and is completed in December, and the current month is May, the progress rate for the construction period = 5/12 ≒ 42% (decimals are rounded off).
- In the case of percentage-of-completion sales, which generally refers to the sales amount of the construction work, the amount of sales of the construction work = the amount of sales of the construction work adjusted to the cumulative progress rate.
・Construction progress rate (cumulative progress rate) refers to the actual progress rate for each construction project. The progress rate generally refers to the degree to which current costs have accumulated compared to the total of the current accumulated costs and the future forecast (the estimated cost amount to be incurred until the construction is completed). For example, if the final cost budget is 1,000 yen and the current incurred costs are 500 yen, then the cumulative progress rate = 500 / (500 + 1,000) ≒ 33%.

From here on, the explanation will mainly focus on graphs, so to make the timeline easier to understand, I will use the term cumulative progress rate.

FIG. 3 is a diagram for explaining the applicable range (applicable industries) of the present invention. In this specification, an embodiment for the construction industry will be described as an example. The present invention is not limited to the building industry,
This is applicable to other companies and industries that manage projects. Figure 3 shows the relationship between the concepts by term in this specification when considering expansion to industries other than the construction industry (software development, advertising/media, etc.).

"Projects (work)" includes projects, which further include construction work, development tasks, and content creation. "Specific conditions" includes project type/duration, which includes work type/construction period/public/private classification, system type/development period, content medium/production period, etc. "Project progress" includes project progress, which includes completed volume, completed man-hours, and completed amount, etc.

As such, the anomaly detection system of the present invention is not limited to the construction industry, but can also be applied to other industries that manage projects, for example.

(1-2. Measures and effects for addressing issues)
The measures and effects for the problems (1) to (4) will be explained with reference to FIGS. 4 to 7.

(Challenge (1))
4 is a diagram for explaining the measures and effects for problem (1). The measures and effects for problem (1) will be explained with reference to FIG.

When the progress rate changes for specific units such as construction period or type of work, it is almost impossible to analyze the progress rate and abnormalities for each unit. When calculating the progress rate under specific conditions, the following issues arise:
- Because it is necessary to define specific conditions and aggregate data from a huge amount of data, there is a risk of calculation errors or data reference errors, which are computationally expensive.
・There are few cases where the "specific conditions that are known to cause different characteristics of output" are known in advance, so it is necessary to separate into multiple units and check the aggregated data to identify the characteristics. Therefore, there are cases where the benefits obtained are low compared to the work costs incurred.

As a solution to this problem, we will automatically calculate the volume per specific condition, and by simply changing the specified specific condition, it will be possible to check whether the specified conditions are "specific conditions known to have different volume characteristics."

This makes it easier to analyze which units have different characteristics when it comes to the volume of work completed by your company, and allows you to quickly narrow down your search to projects with similar volumes, allowing you to more accurately identify abnormalities in volume.

(Challenge (2))
5 is a diagram for explaining the measures and effects for problem (2). The measures and effects for problem (2) will be explained with reference to FIG.

It is not realistic, and is virtually impossible, to regularly check the progress trends of all past construction projects. When trying to refer to the progress records of all construction projects, the following issues arise:
- It is necessary to extract information about production volume from a huge amount of data, which is costly and prone to errors.
- Calculating volume trends requires specialized skills, so in addition to understanding data structure, it is also necessary to calculate volume over time.

As a result, regular checks are required to understand trends, and the two risks mentioned above arise each time.

As a countermeasure, the system periodically checks the progress of all past construction projects, and performs calculations on the system side to visualize the progress of all past construction projects. In addition, a graph is provided on the analysis screen to visualize the progress of progress.

This allows you to see the progress of past construction work with just one graph. This reduces work costs and the risk of errors, and eliminates the need for specialized skills to visualize progress.

(Challenge (3))
6 is a diagram for explaining the measures and effects for problem (3). The measures and effects for problem (3) will be explained with reference to FIG.

In order to compare the trends in progress volume of construction work with the trends in volume of all past construction work, it is necessary to confirm the approximate value of the volume of work completed during the construction period for all past construction work. When trying to obtain an approximate value from the trends in volume of all construction work, the following issues arise:

- It is necessary to extract information about production volume from a huge amount of data, which is costly and prone to errors.
- Calculating volume trends requires specialized skills, so in addition to understanding data structure, it is also necessary to calculate volume over time.
- Calculating an approximate value of trading volume requires specialized skills, so calculations are required to calculate an approximate value of the data taking into account the characteristics of the data trends.

As a countermeasure, the system's algorithm is used to calculate an approximate value of the progress made from all past construction work, and the approximate value is visualized on an analysis screen.

This allows users to intuitively grasp trends in the volume of work completed over time for past construction projects, enabling them to confirm approximate trends in the volume of work completed over time for their own projects, taking into account the characteristics of all past projects. This reduces work costs and the risk of errors, and eliminates the need for specialized skills to calculate approximate data values.

(Challenge (4))
7 is a diagram for explaining the measures and effects for problem (4). The measures and effects for problem (4) will be explained with reference to FIG.

In order to notice that the progress of a construction project in progress is abnormal compared to past construction projects, it is necessary to compare each individual progress trend point (each month, 10% progress intervals, etc.) with the approximate value of all past construction projects, which is unrealistic and virtually impossible. When trying to compare each individual progress trend point (each month, 10% progress intervals, etc.) with the approximate value of all past construction projects, the following issues arise:

- It is necessary to compile data by pattern, such as by aggregating the volume of work completed at each point in the progress (each month, 10% intervals of progress, etc.), and because this requires the task of compiling data by pattern from a huge amount of data, in addition to incurring work costs and the risk of errors, knowledge of pattern compilation is required. Furthermore, since it is necessary to compare the results of all work in progress and determine that any values that deviate abnormally from the approximate values are abnormal, it is virtually impossible to perform this task manually.

As a countermeasure, the "progression of progress for construction work in progress" and the "approximate progress of progress for all past construction work" are visualized side by side. It is also possible to check the progress rate and month by month comparisons on a single screen. The system also calculates and visualizes a threshold that can be used to determine an abnormality based on the deviation from the approximate value. Furthermore, it visualizes whether construction work in progress is showing an abnormal progress rate trend at a glance.

This makes it possible to immediately grasp deviations from past trends at each progress rate point. This reduces work costs and the risk of errors, eliminates the need for manual data compilation, and makes it possible to check and understand at a glance whether the data is abnormal.

(1-3. Analysis screen)
The analysis screen of the anomaly detection system of this embodiment will be described with reference to Figures 8 and 9. The analysis screen outputs "progress rate and cumulative progress rate by construction project,""approximate cumulative progress rate by progress rate for the entire construction project," and "forecasted sales amount (approximate value) and actual results by organization and construction project." Information for which an anomaly has been detected is highlighted by changing the color and font size so that the anomaly can be identified at a glance. The anomaly detection data that can be checked targets the cumulative progress rate by progress rate for construction projects in progress, and detects construction projects that deviate significantly from the trend of past cumulative progress rates.

The analysis screen consists of an initial analysis screen as shown in Figure 8 and an analysis screen as shown in Figure 9. The analysis screen switches from Figure 8 to Figure 9 in order.

The initial analysis screen in Figure 8 displays a list of detected anomalies and outputs an overview of the detected anomalies.

Area A1 displays messages related to the anomaly detection process, showing the difference between the cumulative progress rate for the timing, construction, and month in which the anomaly was detected and the approximate progress rate for past construction work. Because the output is on an overview basis, you must select a message and switch screens to view detailed detection methods. If multiple anomalies are detected, multiple messages will be displayed vertically.

The analysis screen in Fig. 9 displays graphs necessary for analyzing messages resulting from detected abnormalities. It switches when a message is selected on the initial analysis screen in Fig. 8. The graphs necessary for analysis are displayed, and the message outputs detailed information related to the detection. The analysis screen in Fig. 9 has multiple display areas, area A1 to area A6.

Area A1 is an area that displays messages related to the anomaly detection process. It displays the "detection method used to detect the anomaly" and "key information for the data that detected the anomaly." If multiple pieces of information are detected as an anomaly, a scroll bar (not shown) is displayed in area A1, and different messages can be displayed by scrolling, even for the number of messages that are not displayed. Clicking on a displayed message will highlight the portion of the graph in areas A2 to A5 that is related to the anomaly detection information that was clicked on.

Area A2 is an area that displays graphs that output "Actual cumulative progress rates for past construction projects and their approximate curves" and "Actual cumulative progress rates linked to construction projects for which an abnormality was detected." It outputs a graph that allows you to check the "trend in cumulative progress rates by construction period progress rate." You can check the plot of cumulative progress rates by construction period progress rate for all past construction projects. You can check the approximate curve that passes through the approximate value calculated from the cumulative progress rate for each construction period progress rate for all past construction projects (this can be considered as the trend in progress rates for your company's construction projects in general). You can check how construction projects for which an abnormality was detected are progressing abnormally far from the approximate curve.

Area A3 displays a graph that outputs the "approximate curve of cumulative progress rate results for past construction projects" and the "cumulative progress rate results linked to construction projects for which abnormalities have been detected" on a monthly basis. It outputs a graph that allows you to check the "trend in cumulative progress rate occurrence by fiscal year and month." By checking the deviation from the approximate curve for each month, you can analyze which months are beginning to show abnormal trends, etc.

Area A4 is an area that displays a table that outputs monthly construction performance and forecast results by organization. The forecast values output in the table that allows you to check "current construction performance and forecast results by organization" are displayed using approximate values calculated from past construction performance. Deviations from the current situation can be checked using a prediction that "approximately, it will end up being around this amount." If progress is poor, it is expected that a trend toward poor performance on an organizational level will be visualized.

Area A5 is an area that displays a table that outputs monthly construction results and forecast results for each project. It outputs a table that allows you to check "current construction results for each project and forecast results expected to occur by the end of the period." Forecast values are displayed using approximate values calculated from past construction results, and a prediction that "approximately, it will end up being around this amount" is used to check for deviations from the current situation. If progress is poor, it is expected that a trend toward poor performance on a project-by-project basis will be visualized. Note that if performance is poor on an organizational level when checked in area A4, it is possible that poor performance across multiple projects will be visualized here.

Area A6 is an extraction condition specification area for specifying extraction conditions such as the base date, accounting period, business establishment, department, and data reference unit for the graphs and tables to be output in areas A2 to A5.
This section is used when you want to check the data extracted based on conditions from the data output to the graphs and tables in areas A2 to A5. When extracting and analyzing the accounting period to be checked, the graphs and tables are output only for construction work that has occurred within the accounting year and month range within the specified accounting period. When extracting and analyzing output data at the organizational level, you can specify the business office or department and output the graphs and tables.

2. Configuration
Fig. 10 is a block diagram showing an example of the configuration of an anomaly detection system 100 according to this embodiment. In Fig. 10, the anomaly detection system 100 includes a control unit 102, a communication interface unit 104, a storage unit 106, and an input/output interface unit 108. The units included in the anomaly detection system 100 are connected to each other so as to be able to communicate with each other via any communication path.

The communication interface unit 104 communicatively connects the anomaly detection system 100 to the network 300 via a communication device such as a router and a wired or wireless communication line such as a dedicated line. The communication interface unit 104 has the function of communicating data with other devices via the communication line. Here, the network 300 has the function of connecting the anomaly detection system 100 and the server 200 etc. so that they can communicate with each other, and is, for example, the Internet or a LAN (Local Area Network).

An input device 112 and an output device 114 are connected to the input/output interface unit 108. The output device 114 may be a monitor (including a home television), a speaker, or a printer. The input device 112 may be a keyboard, a mouse, a microphone, or a monitor that functions as a pointing device in conjunction with a mouse. Note that, below, the output device 114 may be referred to as the monitor 114.

The storage unit 106 stores various databases, tables, files, and the like. The storage unit 106 also stores computer programs that work in conjunction with the OS (Operating System) to issue commands to the CPU (Central Processing Unit) to perform various processes. The storage unit 106 can be, for example, a memory device such as RAM (Random Access Memory) or ROM (Read Only Memory), a fixed disk device such as a hard disk, a flexible disk, or an optical disk.

The memory unit 106 includes a business database 106a, an anomaly detection execution data table 106b, an anomaly determination result data table 106c, an anomaly determination definition master 106d, etc.

The business database 106a is a database for storing business data (also called "project data"), accounting period master data, etc. Construction data will be explained as an example of business data. Other business data includes development task data, production content data, etc.

Construction data may include the name of the construction project, type of work (specific conditions), organization (business establishment, department), construction start month, construction end month, construction period (specific conditions), construction completion status (completed or incomplete), accounting year and month, sales amount, construction period progress rate, cumulative progress rate, etc. (see Figure 12 (A), etc.).

The accounting period master can be composed of a table that associates and registers accounting periods, accounting years and months, and display order (see Figure 23A (D)).

The anomaly detection execution data table 106b is a table for storing automatic detection execution schedule data, acquisition range condition data for business data (e.g., construction data), and data on the period for registering anomaly determination results.

The automatic detection execution schedule data may include a detection ID, schedule ID, execution conditions, and execution time (see Figure 13). The detection ID and schedule ID serve as key information when referencing the data. The detection unit 102b automatically executes anomaly detection for construction data in accordance with the automatic detection execution schedule data.

The acquisition range condition data for business data (e.g., construction data) may include a detection ID, schedule ID, condition, and condition value (see Figure 14 (C)). The detection ID and schedule ID serve as key information when referencing the data. The detection unit 102b divides the population of construction data for each specific condition (e.g., type of work) and automatically performs anomaly detection for each population.

The period data for which anomaly determination results are registered may include a detection ID, period determination items, and acquisition criteria (see Figure 13). The detection ID serves as key information when referencing the data. The detection unit 102b sets the construction data for period determination items (e.g., fiscal year and month) that match the acquisition criteria as targets for anomaly determination and performs an anomaly determination.

The abnormality determination result data table 106c is a table for storing the results of abnormality detection execution, such as abnormality determination result data, abnormality determination result message data, detailed abnormality determination result message data, and ancillary information data for the abnormality determination result. When the detection unit 102b detects an abnormality in the construction data, it stores the abnormality determination result data, abnormality determination result message data, detailed abnormality determination result message data, and ancillary information data for the abnormality determination result data table 106c.

The abnormality determination result data may include the detection ID, JOB ID, determination group CD, message ID, fiscal year and month, business establishment, department, construction project, type of work, abnormality determination result (True if abnormal, False if not abnormal), construction progress rate, cumulative progress rate, predicted progress rate (predicted cumulative progress rate), etc. (See Figure 19A (A)). The "determination group CD" is the key for the determination group (population). The detection ID, JOB ID, and determination group CD become key information when referencing the data. If the determination result indicates that there is no abnormality, no determination result message data record is created.

The abnormality determination result message data may include the detection ID, JOB ID, determination group CD, message ID, abnormality level, definition name, summary, and detection target (see Figure 19A (B)). The detection ID, JOB ID, and determination group CD serve as key information when referencing the data.

The detailed data of the abnormality determination result message may include the detection ID, JOB ID, determination group CD, message ID, detection technique, determination method, lower limit value, upper limit value, etc. (See Figure 19B (C), etc.). The detection ID, JOB ID, and determination group CD serve as key information when referencing the data.

The anomaly detection result ancillary information data may include the detection ID, JOB ID, judgment group CD, line number, construction progress rate, predicted progress rate, lower limit, upper limit, etc. (See Figure 19B (D)). The detection ID, JOB ID, and judgment group CD serve as key information when referencing the data.

The anomaly determination definition master 106d can be configured as a table that associates and registers the anomaly determination definition ID, anomaly determination definition name, acquisition definition, algorithm used (e.g., approximate curve), and parameter settings (X-axis item (e.g., progress rate of construction), Y-axis item (e.g., cumulative progress rate), X-axis scale unified item, Y-axis scale unified item, approximate function candidate name (e.g., polynomial function, logistic function, Bass model function), smoothness, significance level)). The operator can set the data in the anomaly determination definition master 106d. The detection unit 102b references the anomaly determination definition master 106d when calculating the approximate curve, predicted cumulative progress rate, and upper and lower limits.

The control unit 102 is a CPU or the like that provides overall control of the anomaly detection system 100. The control unit 102 has internal memory for storing control programs such as an OS, programs that define various processing procedures, and required data, and performs various information processing operations based on these stored programs.

The control unit 102 is configured to be able to access the business database 106a, the data table for abnormality detection execution 106b, the data table for abnormality determination results 106c, the abnormality determination definition master 106d, and other data stored in the memory unit 106. Note that the business database 106a, the data table for abnormality detection execution 106b, the data table for abnormality determination results 106c, and the abnormality determination definition master 106d may be provided in another location (e.g., server 200) as long as they are accessible by the control unit 102.

Functionally, the control unit 102 comprises a memory control unit 102a, a detection unit 102b, and a display control unit 102c.

The storage control unit 102a may, for example, acquire daily business data from a business system (not shown) connected via the network 300 and store it in the business database 106a, or may input daily business data in response to an operator's operation on an input screen (not shown) displayed on the monitor 114 and store it in the business database 106a.

In addition, the memory control unit 102a sets automatic detection execution schedule data, construction data acquisition range condition data, and abnormality determination result registration period data, for example, in response to operator operations on a data setting screen (not shown) displayed on the monitor 114, and stores these in the abnormality detection execution data table 106b.

The detection unit 102b performs anomaly detection in accordance with the automatic detection execution schedule data, business data acquisition range condition data, and anomaly determination result registration period data stored in the anomaly detection execution data table 106b. Based on the business data stored in the business database 106a, the detection unit 102b divides the population of past business data into specified specific conditions, calculates an approximate curve for the business data progress rate and project progress rate for each population, calculates a predicted project progress rate by substituting the progress rate into the approximate curve function, calculates an upper limit (also referred to as the "upper limit threshold") and a lower limit (also referred to as the "lower limit threshold") based on the approximate curve and the predicted project progress rate, and detects an anomaly by comparing the business data to be detected with the upper limit and lower limit. If the detection unit 102b detects an anomaly, it stores the anomaly determination result data, anomaly determination result message data, anomaly determination result message detail data, and anomaly determination result ancillary information data in the anomaly determination result data table 106c.

The display control unit 102c controls the display of analysis screens (initial analysis screen, analysis screen) on the monitor 114. Based on the business data stored in the business database 106a, the display control unit 102c displays analysis data (e.g., Graphs 1-2, Tables 3-4 in Figure 4, etc.) on the analysis screen for business data in which an abnormality has been detected by the detection unit 102b.

The business data includes construction data, the specific conditions include the type of work, construction period, or public/private classification, the progress rate includes the construction period progress rate calculated by dividing the number of months elapsed in construction by the planned construction period, and the project progress rate includes the cumulative progress rate calculated by dividing the currently incurred costs by (budgetary cost + currently incurred costs), and the construction data may include the construction, organization (business location, department), one or more specific conditions, fiscal year and month, sales amount, construction period progress rate, and cumulative progress rate.

Based on the construction data, the display control unit 102c may display a message in a specified area of the analysis screen (e.g., area A1 in Figure 9) that includes the construction work detected as abnormal, the detection method, and the cumulative progress rate at which the abnormality was determined.

Based on the construction data, the display control unit 102c may display in a specified area of the analysis screen (e.g., area A2 in Figure 9) the actual cumulative progress rate for past construction projects by population or for all past construction projects, broken down by construction period progress rate, and its approximate curve, as well as a graph showing the trend in the actual cumulative progress rate associated with construction projects in which an abnormality was detected.

Based on the construction data, the display control unit 102c may display, in a specified area of the analysis screen (e.g., area A3 in Figure 9), the actual cumulative progress rate of past construction work by accounting year and month, its approximate curve, and a graph showing the trend in the actual cumulative progress rate linked to construction work in which an abnormality was detected.

Based on the construction data, the display control unit 102c may display a table in a specified area of the analysis screen (e.g., area A4 in Figure 9) showing trends in construction sales actuals (construction actuals) and sales forecasts (construction forecasts) by organization within a specified accounting period.

The display control unit 102c may display, based on the construction data, a table in a specified area of the analysis screen (e.g., area A5 in Figure 9) showing the actual sales performance and sales forecast trends for each construction project of the organization in charge of the construction project in which the abnormality occurred.

3. Specific Examples
Specific examples of processing by the control unit 102 of the anomaly detection system 100 in this embodiment will be described in the following order with reference to FIGS. 10 to 74: [3-1. Overall processing], [3-2. Sample data], [3-3. Methods for calculating approximate curves, predicted values, and upper and lower limit values], and [3-4. Options for processing methods for past cases].

[3-1. Overall processing]
FIG. 11 is a flowchart for explaining an outline of the overall processing of the control unit 102 of the anomaly detection system according to this embodiment.

Referring to FIG. 11, an overview of the overall processing of the control unit 102 of the anomaly detection system 100 in this embodiment will be described. In FIG. 11, the detection unit 102b executes the anomaly detection process (step S1). Specifically, in the anomaly detection process, the detection unit 102b executes anomaly detection for construction data in accordance with the automatic detection execution schedule data, business data acquisition range condition data, and anomaly determination result registration period data stored in the anomaly detection execution data table 106b. Based on the construction data stored in the business database 106a, the detection unit 102b divides the population of past construction data into specified specific conditions, calculates an approximate curve for the construction data's construction period progress rate and cumulative progress rate for each population, calculates a predicted value for the cumulative progress rate by substituting the construction period progress rate into the approximate curve function, calculates upper and lower limit values based on the approximate curve and the predicted value of the cumulative progress rate, and detects anomalies by comparing the construction data to be detected with the upper and lower limit values. If the detection unit 102b detects an abnormality, it stores the abnormality determination result data, abnormality determination result message data, abnormality determination result message detailed data, and abnormality determination result ancillary information data in the abnormality determination result data table 106c.

The display control unit 102c executes the analysis screen display process (step S2). Specifically, in the analysis screen display process, the display control unit 102c controls the display of the analysis screen (initial analysis screen, analysis screen) on the monitor 114. Based on the construction data stored in the business database 106a, the display control unit 102c displays analysis data (e.g., the message, graphs 1-2, tables 3-4, etc. in Figure 9) on the analysis screen for the construction work in which the detection unit 102b detected an abnormality.

In addition, the display control unit 102c may display a message in a specified area of the analysis screen (e.g., area A1 in Figure 9) based on the construction data, including the construction work detected as abnormal, the detection method, and the cumulative progress rate at which the abnormality was determined.

In addition, based on the construction data, the display control unit 102c may display in a specified area of the analysis screen (for example, area A2 in Figure 9) the actual cumulative progress rates and their approximate curves for past construction projects by population or for all past construction projects, broken down by construction period progress rate, as well as a graph showing the trends in the actual cumulative progress rates associated with construction projects in which an abnormality was detected.

In addition, the display control unit 102c may display, in a specified area of the analysis screen (e.g., area A3 in Figure 9) based on the construction data, the actual cumulative progress rate of past construction work by accounting year and month, its approximate curve, and a graph showing the trend in the actual cumulative progress rate linked to construction work in which an abnormality was detected.

The display control unit 102c may also display a table showing the actual sales figures and forecasted sales trends for construction work by organization within a specified accounting period in a specified area of the analysis screen (e.g., area A4 in Figure 9) based on the construction work data.

In addition, the display control unit 102c may display a table showing the trends in actual sales and sales forecasts for each construction project of the organization in charge of the construction project in which the abnormality occurred in a specified area of the analysis screen (e.g., area A5 in Figure 9) based on the construction project data.

[3-2. Sample data]
12 to 51 are diagrams showing sample data for explaining a specific example of the processing by control unit 102 of anomaly detection system 100 according to this embodiment. A specific example of the processing by control unit 102 of anomaly detection system 100 according to this embodiment will be described with reference to FIGS.

(S1: Abnormality detection process)
12 to 19, a specific example of anomaly detection processing will be described. The detection unit 102b executes anomaly detection for construction data in accordance with the automatic detection execution schedule data, business data acquisition range condition data, and anomaly determination result registration period data stored in the anomaly detection execution data table 106b, divides a population of past construction data for each specified specific condition based on the construction data stored in the business database 106a, calculates an approximate curve for the construction data's progress rate and cumulative progress rate for each population, calculates a predicted value for the cumulative progress rate by substituting the construction data's progress rate for the approximate curve function, calculates upper and lower limit values based on the approximate curve and the predicted value for the cumulative progress rate, and detects anomalies by comparing the construction data to be detected with the upper and lower limit values.

(Terminology)
・The progress rate for the construction period refers to the progress rate for the construction period as of the current accounting month and year in months. For example, if we assume that there is a construction project with a 12-month construction period, which begins in January and is completed in December, and the current month is May, the progress rate for the construction period = 5/12 ≒ 42% (decimals of % are rounded up or down).
- Cumulative progress rate refers to the actual progress rate for each construction project. The progress rate generally refers to the degree to which current costs have accumulated compared to the total of the current accumulated costs and the estimated final cost (the estimated cost amount to be incurred until the construction is completed). For example, if the final cost budget is 1,000 yen and the current incurred cost is 500 yen, then the cumulative progress rate = 500 / (500 + 1,000) ≒ 33%.
The predicted progress rate is a predicted value of the progress rate at an approximate position based on the past construction results, as determined by the abnormality detection process of the detection unit 102b.

The method for calculating the approximate curve, forecast value, upper and lower limit values from the construction data's construction progress rate and cumulative progress rate is explained in detail in (3-3. Method for calculating the approximate curve, forecast value, upper and lower limit values), and only an outline will be explained here.

1. A process that calculates the predicted construction progress rate for all managed construction projects based on construction data in the business database 106a, and detects construction projects that are progressing abnormally.

Figure 12 is a diagram explaining construction data. Construction data is used to store information about construction work being carried out (organization, construction period, start date, completion date, planned man-hours, order amount, cumulative progress rate, etc.), and as shown in Figure 12 (A), includes fields such as construction name, business location, department, construction start month, construction end month, construction period, construction completion category, accounting year and month, sales amount, construction period progress rate, and cumulative progress rate.

In the construction industry, companies that use percentage-of-completion sales often already manage the progress rate and cumulative progress rate in their business data. Therefore, this specification will explain the situation where the progress rate and cumulative progress rate already exist in the business data. In the construction industry, companies that use sales based on completed work or companies that manage projects in other industries may not manage the progress rate, in which case a separate calculation is required.

The "construction progress rate" is calculated using the following formula.
"Construction progress rate" = (number of months elapsed since construction began ÷ total number of months required for construction × 100
= (Number of months from "Construction start month" to "Fiscal year and month" + 1 ÷ Number of months in "Construction period") x 100

The "fiscal year and month" in which the above construction projects A, B, and C were carried out can be summarized as shown in the table in Figure 12 (B). The numbers in the colored cells represent the "construction progress rate."

(1) Pre-setting (1-1) The storage control unit 102a stores information necessary for anomaly detection in the anomaly detection execution data table 106b in advance (data provided in advance). Specifically, automatic detection execution schedule data and anomaly determination result registration target period data are set in the anomaly detection execution data table 106b.

Figure 13 (A) is a diagram showing an example of automatic detection execution schedule data. Automatic detection execution schedule data has fields for detection ID, schedule ID, execution conditions, and execution time. In the example shown in the figure, the detection ID is "AB001," the schedule ID is "SH001," the execution condition is "5th business day of each month," and the execution time is "23:00." In this example, automatic detection is executed at "23:00" on the 5th business day of each month. The 5th business day is assumed to be the time when the monthly closing process for the previous month has been finalized.

Figure 13 (B) shows an example of data for the period for which anomaly determination results are registered. The data for the period for which anomaly determination results are registered includes the following fields: detection ID, period determination item, and acquisition criteria. In the example shown in the figure, the detection ID is "AB001," the period determination item is "fiscal year and month," and the acquisition criteria item is "the month preceding the month in which the startup date falls." In this example, the acquisition criteria is "the month preceding the month in which the startup date falls," so for example, if startup occurs on 10/5/2022, anomaly determination will be performed on the data for 9/2022. Detailed internal setting items are not directly related to this application, so they will be omitted here.

(2) The detection unit 102b automatically executes anomaly detection.
(2-1) First, acquire timing information for detecting an abnormality. Specifically, acquire automatic detection execution schedule data and abnormality determination result registration target period data set in the abnormality detection execution data table 106b in (1-1).

(2-2). Determine whether the timing of automatic execution is the timing to detect an abnormality. Figure 14 (A) is a diagram for explaining the automatic execution timing determination process. In the example shown in the figure, the start timing is "2022/11/07", the fifth business day, and the automatic detection execution schedule data is "the fifth business day of every month", so the execution determination result is "execute."

The execution decision determines whether or not to perform subsequent processing. If the execution decision result is "execute," then (2-3) and subsequent steps are executed. If the execution decision result is "do not execute," then processing ends.

Business days are determined by referencing the calendar master (not shown) in the business database 106a to determine whether they are holidays or business days. The calendar master is based on the sales business calendar and is always up to date.

(2-3) Obtain the range conditions for the data to detect anomalies.
Specifically, for example, the detection ID "AB001" and schedule ID "SH001" of the automatic detection execution schedule data shown in FIG. 14(B) are used as parameters (keys) to acquire the acquisition range condition data for construction data set in the anomaly detection execution data table 106b as shown in FIG. 14(C). The acquisition range condition data for construction data has items for detection ID, schedule ID, condition, and condition value. In the example shown in the figure, the items are detection ID "AB001", schedule ID "SH001", condition "data reference unit", and condition value "type of construction".

The "condition value" can be set in a general way depending on the company using it. In many cases, the shape of the progress rate change will differ depending on the characteristics of the construction work, such as the type of work. Construction characteristics that similarly show changes in progress rate include construction period, type of work, and public/private classification (public works or private works). Here, as an example, we will explain the pattern set for type of work. Note that all of the above characteristics are assumed to be information contained in the construction data, just like the type of work. It is also possible to set the condition value as empty. If empty, it is possible to analyze changes in progress rate for all construction work as a single unit.

(2-4) The construction data in the business database 106a is referenced to calculate the predicted progress rate for each construction project and construction period progress rate.
(2-4-1) Construction data is acquired from the business database 106a. The values within the acquisition range condition data of the construction data acquired in (2-3) are set as parameters to acquire the data. For example, using the parameters (data reference unit, work type) as shown in FIG. 15(A) as keys, the construction data in the business database 106a is referenced and populations are divided and acquired by data reference unit. For example, construction data for population A: work type (construction) and population B: work type (demolition) as shown in FIGS. 15(B) and (C) is acquired. An approximate value is calculated for each population.

(2-4-2). Using the construction data obtained in (2-4-1), calculate the cumulative progress rate (approximate value) by construction period progress rate for each population. Identify abnormal construction work based on the upper and lower limits considered to be the normal range obtained when calculating the approximate value. The "approximate value" is a reference value that will be used to compare with actual results, and will hereafter be referred to as the "predicted value."

Figure 16 (A) shows the acquired construction data, and Figure 16 (B) plots each construction project in the construction data, with the horizontal axis showing the construction progress rate (%) and the vertical axis showing the cumulative progress rate (%).

The "construction period progress rate" and "cumulative progress rate" in the construction data are used to calculate predicted cumulative progress rates by construction period progress rate. The method for calculating predicted values is to use an anomaly detection algorithm: approximation curve. The data used for the approximation curve is limited to construction data for which the "construction completion category" is complete. The predicted progress rate by month and construction project is also calculated using the approximation formula for the approximation curve used to calculate the predicted value.

Figure 17 is a diagram for explaining the calculation of the approximate curve F(x), predicted progress rate, and upper and lower limits. As shown in Figure 17(A), the completed construction work is used as the population, and the approximate curve F(x) of the cumulative progress rate by construction period progress rate for the population is calculated. In addition, the standard deviation function S(x) is calculated by focusing on the tendency of data dispersion within the population. Then, as shown in Figure 17(B), the construction period progress rate is substituted into the approximate curve F(x) and the standard deviation function S(x), respectively, to calculate the predicted progress rate and upper and lower limits for the uncompleted construction work. In this way, the predicted progress rate is calculated using F(x), and the upper and lower limits are calculated using S(x).

The key to this invention is using the results calculated by this process, so the specific calculation process is explained in detail in [3-3. Calculation methods for approximation curves, forecast values, and upper and lower limits], and only an outline will be explained here. This explanation targets companies with percentage-of-completion sales, which means that the construction data already contains the construction progress rate and cumulative progress rate, eliminating the need to scale the approximation curve or process the data.

For example, predicted values and upper and lower limit values are calculated for each work type based on the approximate curve parameters (X-axis item: construction progress rate, Y-axis item: cumulative progress rate, approximate function candidates [polynomial function, logistic function, Bass model function], etc.) as shown in Figure 18A (A) set in the anomaly judgment definition master 106d. Note that since the explanation this time uses values that have already been scaled, an explanation of scale adjustment will be omitted. After calculating F(x), the data dispersion is calculated to calculate the standard deviation function S(x). The construction progress rate is substituted for x in F(x) and S(x), and the predicted value is calculated for F(x) and the upper and lower limit values are calculated for S(x). The same function is used to calculate the predicted value and upper and lower limit values for each work type and construction project.

Figure 18A (B) shows an example of approximate data for cumulative progress rates calculated by construction and demolition period progress rates, which are calculated by population, and includes the following fields: work, type of work, construction period progress rate, cumulative progress rate, forecast value (forecasted progress rate, forecasted value of cumulative progress rate), lower limit, and upper limit.

Figure 18B (C) shows the construction data after the predicted value and upper and lower limits have been calculated. In addition to the predicted value (predicted progress rate), lower limit, and upper limit, it also stores information on the construction type, which is the data reference unit.

According to the "acquisition criteria" for the anomaly determination result registration target period data acquired in (1-1) shown in Figure 18B (D), a determination is made as to whether there are any anomalies in the data for the month prior to the month to which the anomaly determination execution date belongs. In this case, the execution date is 2022/11/07, so a determination is made as to whether there are any anomalies on 2022/10. In this example, the record on the fourth line in Figure 18B (C) has a cumulative progress rate of "30%," a predicted progress rate of "65%," a lower limit of "60%," and an upper limit of "75%." Because the cumulative progress rate of "30%" is below the lower limit of "60%, an anomaly is detected.

(2-4-3). Works determined to be abnormal in (2-4-2) are saved in the abnormality determination result data table 106c. At the same time, a message to be displayed on the analysis screen is saved. Figures 19A and 19B (A) to (D) show examples of abnormality determination result data, abnormality determination result message data, detailed abnormality determination result message data, and abnormality determination result ancillary information data. Note that the abnormality determination result ancillary information data holds display data for each population obtained. For results that are determined not to be abnormal as a result of the determination, no determination result message record is created.

(S2: Analysis screen display process)
The analysis screen display process will be described in detail with reference to Figures 20 to 51. The display control unit 102c displays analysis data (e.g., the message, graphs 1 and 2, tables 3 and 4 in Figure 9, etc.) on the analysis screen for construction work for which an abnormality has been detected by the detection unit 102b, based on the construction work data stored in the business database 106a.

2. Execute a process to display the abnormal data detected in 1 and data related to the abnormal data on the initial screen for analysis.
(1) Information automatically detected as an anomaly from the construction data in the business database 106a is output to the initial screen for analysis.
(1-1) First, obtain the result data and result message of the detected abnormality.
Specifically, as shown in FIG. 20A(A), using the JOBID "output forecast alert" and the abnormality determination result "True" as parameters (keys), the abnormality determination result data as shown in FIG. 20A(B), the abnormality determination result message data as shown in FIG. 20B(C), and the detailed abnormality determination result message data as shown in FIG. 20B(D) are obtained from the abnormality determination result data table 106c. Note that in order to display the detected information in a list, data is obtained without the detection ID being set. On the screen that will be displayed from now on, only data for which the abnormality determination result has been determined to be abnormal will be checked. Therefore, only data for which the value of the "Abnormal result or not" column in the abnormality determination result data is "True" will be referenced.

(1-2) When an abnormality is detected, a message is displayed on the screen.
Specifically, the abnormality level, definition name, summary, and detection target of the abnormality determination result message data in FIG. 20B(C) are extracted and displayed in area A1 of the initial analysis screen as shown in FIG.

(1-3). The date when the initial analysis screen is launched is obtained and set as the base date for the extraction conditions on the initial analysis screen. Specifically, as shown in Figure 21, the base date "2022/11/1" is set as the base date for the extraction conditions on the initial analysis screen.

(2) Select the result message for the detected abnormality on the initial analysis screen and launch the analysis screen. The following process is performed by referencing the abnormality judgment result data, abnormality judgment result message data, and abnormality judgment result message detailed data selected from the list in (1).

(2-1) Switch the message to detailed information display and secure the output area for graphs and tables.
1. Switching to a detailed message display. Specifically, for the detailed data of the abnormality determination result message shown in Fig. 22A (A), the detection method, determination method, and lower limit value are referenced, and a detailed message display such as that shown in Fig. 22A (B) is performed (the detection method and the reason for determination are added).

2. Reserve an output area for graphs and tables on the analysis screen. In this system, for example, as shown in Figure 22B (C), figures 1 to 4 (two graphs and two tables) are output. Therefore, an output area large enough for two graphs and two tables is reserved.

(2-2) Set the extraction conditions.
1. Set extraction conditions for the accounting year and month of the abnormality determination result data linked to the abnormality determination result message data.

(i) From the accounting year and month of the target data, obtain the accounting period that includes that accounting year and month from the accounting period master, and set the beginning and ending accounting years and months associated with the accounting period as extraction conditions for the analysis screen.

Figure 23A (D) shows an example of accounting period master data, which has fields for accounting period, accounting year and month, and display order. The accounting period master manages the accounting years and months included in each accounting period. For example, using the accounting year and month "2022/10" of the anomaly determination result data shown in Figure 23A (C) as a key, accounting period "50" including "2022/10" is obtained from the accounting period master shown in Figure 23A (D), and the beginning and ending accounting years and months "2022/04-2023/03" linked to accounting period "50" are then obtained and set as the initial values in the extraction conditions, as shown in Figure 23A (A).

(ii) The fiscal year and month of the beginning and end of the fiscal year obtained from the fiscal year master are obtained as the setting values for the extraction conditions and set as the extraction conditions for the analysis data acquisition range conditions. In addition, the construction work that is the source of the selected abnormality is also obtained and set as the setting value for the extraction conditions. Figure 23A (B) shows an example of the analysis data acquisition range conditions, with the extraction condition column being "Fiscal Year and Month," the FROM condition being "2022/04," the TO condition being "2023/03," and the target construction being "Construction A."

2. Set the extraction conditions with the information in 1 set as the initial display values. Specifically, as shown in Figure 23B (A), the initial extraction condition values are set as follows: base date "2022/11/1" (timing when the analysis screen is launched), start of period "2022/04" (FROM condition of the analysis data acquisition range conditions), end of period "2023/03" (TO condition of the analysis data acquisition range conditions), and target construction "Construction A" (target for acquisition of analysis data).

As shown in Figures 23B (B) and (C), when switching to graph display mode, filter items for extracting data for graph output are displayed. Fixed items are output. If a judgment group CD and a judgment group key are set for the abnormality judgment result data, an additional extraction item is displayed so that data can be extracted by selecting the judgment group key. In this example, the construction type "construction" is additionally displayed.

(2-3) Cumulative progress rate transition - actual results data is obtained from the construction data in the business database 106a. Based on the conditions initially set in the extraction conditions, cumulative progress rate transition - actual results data is obtained and used as data for display in Figures 1 and 2.

Specifically, the construction data shown in Figure 24B(D) is extracted based on the parameters (keys) of the start period "2022/04," end period "2023/03," construction "Construction A," and construction type "Construction" as shown in Figure 24A(A), and the cumulative progress rate trend_actual data shown in Figure 24B(C) is obtained. The cumulative progress rate trend_actual data includes items such as construction name, business establishment, department, construction type, construction completion category, accounting year and month, sales amount, construction period progress rate, and cumulative progress rate.

In this case, if there is even one record that includes the "fiscal year and month" in the period between the start and end of the period, that construction project will be eligible for acquisition. For example, as shown in Figure 24A (B), in this example, construction projects A and B are eligible for acquisition, but construction project C is not eligible for acquisition.

(2-4) Obtain "cumulative progress rate trend approximate curve data" from the abnormality determination result data, abnormality determination result message data, and abnormality determination result ancillary information data in the abnormality determination result data table 106c.

Figure 25B (C) shows an example of abnormality determination result data, Figure 25B (D) shows an example of abnormality determination result message data, Figure 25B (E) shows an example of abnormality determination result ancillary information data, and Figure 25B (F) shows an example of cumulative progress rate trend approximate curve data (construction).

Using the start period "2022/04," end period "2023/03," work type "Work A," and work type "Construction" as parameters (keys) as shown in Figure 25A (B), the cumulative progress rate transition approximate curve data (construction) shown in Figure 25B (F) is obtained as additional display data for Figures 1 and 2 from the abnormality determination result data shown in Figure 25B (C), the abnormality determination result message data shown in Figure 25B (D), and the abnormality determination result ancillary information data shown in Figure 25B (E).

The cumulative progress rate trend - approximate curve data (construction) includes the following fields: project name, construction progress rate, predicted progress rate, upper progress rate limit, and lower progress rate limit threshold.

(2-5). Obtain "sales amount trend data by organization" from the cumulative progress rate trend_actual data obtained in (2-3) and the abnormality determination result data in the abnormality determination result data table 106c.

Figure 26B (D) shows an example of the cumulative progress rate trend - actual data acquired in (2-3), Figure 26B (E) shows an example of anomaly determination result data, and Figure 26A (C) shows an example of sales amount trend data by organization.

As shown in Figure 26A (A), using the start period "2022/04," end period "2023/03," and construction type "construction" as parameters (keys), the cumulative progress rate trend-actual data shown in Figure 26B (D) and the abnormality determination result data shown in Figure 26B (E) are used to obtain the sales amount trend data by organization shown in Figure 26A (C) as data for display in Chart 3. Graph 3 checks only the construction results that occurred within the relevant accounting period, and obtains and displays only the data for the results within the box shown in Figure 26A (B).

For cumulative progress rate trend - actual data, if there are results for the previous period, the results excluding the previous period will be obtained (only results that can be counted as for the current period will be obtained). For example, for the cumulative progress rate of project A, 1% for 2022/03 is the previous period's results, 2% for 2022/04 but 1% for the current period, 4% for 2022/05 but 3% for the current period, etc.

As shown in Figure 26A (C), the sales amount trend data by organization includes the following fields: business establishment, department, accounting year/month, type of work, sales amount (actual), and sales amount (forecast). "Sales amount (actual)" and "sales amount (forecast)" are the actual and forecast values of the cumulative recorded amount of "sales amount," respectively, and are calculated using the following formula:

Sales amount (actual) = Sales amount x (cumulative progress rate ÷ 100)
・Sales amount (forecast) = Sales amount x (forecast progress rate ÷ 100)

Create records for all fiscal years and months in the period range (for graph drawing). For fiscal years and months with no actual or forecast results, create records with 0 yen. Obtain records (1) aggregated by business establishment/department and records (2) aggregated by business establishment.

In the example for the fiscal year and month "2022/04", the results are: Department A: Actual results "400", forecast "600", Department B: Actual results "12,000", forecast "12,400" (in this explanation, these records will not be retrieved due to the work type extraction condition "Construction", but for the sake of explaining the summary calculation, we will assume that they were retrieved), Business Office A (Department A + Department B): Actual results "12,400", forecast "13,000".

(2-6) Obtain the sales amount trend data by construction project from the "Cumulative progress rate trend - actual data" obtained in (2-3) and the abnormality judgment result data in the abnormality detection result data table as display data for Figure 4.

Figure 27A (C) shows an example of data on the trend in sales amount by construction project, Figure 27B (D) shows an example of data on the trend in cumulative progress rate - actual results data obtained in (2-3), and Figure 27B (E) shows an example of data on the abnormality determination results.

Using the start period "2022/04," end period "2023/03," work "Work A," and work type "Construction" as parameters (keys) as shown in Figure 27A (A), the sales amount trend data by work shown in Figure 27A (C) is obtained as data for display in Figure 4 from the cumulative progress rate trend - actual data shown in Figure 27B (D) and the abnormality determination result data shown in Figure 27B (E). If there is even one record in the above period that includes the "fiscal year and month," that work will be acquired. For example, in the table shown in Figure 27A (A), works A and B are eligible for acquisition, but work C is not.

Here, extraction is not performed by construction number. Data is extracted and obtained by the business office linked by the construction number. For example, if Construction 1 is a construction project managed by the Tokyo business office, the actual data for the Tokyo business office is obtained. The determination of which office manages the project is made based on whether or not there are actual results in "Cumulative progress rate trend - actual data."

As shown in Figure 27A (C), the sales amount trend data by construction project includes the following fields: project name, business location, department, accounting year and month, construction type, sales amount (actual), sales amount (forecast), and forecast progress rate. "Sales amount (actual)" and "Sales amount (forecast)" are the actual and forecast values of the cumulative recorded amount of "sales amount," respectively, and are calculated using the following formula:

Sales amount (actual) = Sales amount x (cumulative progress rate ÷ 100)
・Sales amount (forecast) = Sales amount x (forecast progress rate ÷ 100)

For the accounting year and month "End of previous period," records before the corresponding accounting period will be aggregated as the previous period. If the accounting year and month at the beginning of the period is 2022/04, records before 2022/03 will be displayed as the previous period. For the accounting year and month "Next period and beyond," records after the corresponding accounting period will be aggregated as the next period and beyond. If the accounting year and month at the end of the period is 2023/03, records after 2023/04 will be displayed as the next period and beyond. Records will also be created if there is no data to be aggregated as the previous period and next period and beyond. In this case, records will be created with actual and forecast values of 0 yen.

(2-7) Bind the cumulative progress rate trend - actual data, cumulative progress rate trend - approximate curve data, sales amount trend data by organization, and sales amount trend data by construction project to their corresponding charts 1 to 4, for a total of four charts.

1. Bind the cumulative progress rate transition_actual data and cumulative progress rate transition_approximate curve data to Chart 1.
For example, in accordance with the binding information (X axis, Y axis, legend, title) shown in Fig. 28A(A) and the data used for each legend shown in Fig. 28A(B), the cumulative progress rate transition_actual data shown in Fig. 28A(C) and the cumulative progress rate transition_approximate curve data shown in Fig. 28A(D) are bound to Chart 1 as shown in Fig. 28B(E). In Fig. 28A(B), it appears that the same plot is used for past performance by construction and performance of construction work determined to be abnormal, but the data is color-coded in the subsequent graph coloring process so that the data can be distinguished.

In Figure 1, the title is "Comparison of forecast/actual production volume curves (construction period progress rate)," the X axis is the construction period progress rate, the Y axis is the cumulative progress rate, and the legend is the forecast production volume approximation curve, the uncompleted work production volume line, the normal data threshold, past performance by construction project, and the performance of construction projects determined to be abnormal.

2. Bind the cumulative progress rate transition_actual data and cumulative progress rate transition_approximation curve data to Chart 2.
For example, in accordance with the binding information (X axis, Y axis, legend, title) shown in FIG. 29A(A) and the data used for each legend shown in FIG. 29A(B), the cumulative progress rate transition_actual data shown in FIG. 29A(C) and the cumulative progress rate transition_approximate curve data shown in FIG. 29A(D) are bound to Chart 2 as shown in FIG. 29B(E).

In Figure 2, the title is Forecast/Actual Volume Curve Comparison (Fiscal Year/Month), the X-axis is the fiscal year/month, the Y-axis is the cumulative progress rate, and the legend is the forecast volume approximation curve, the uncompleted work volume line, the normal data threshold, and Work A.

3. Bind the sales amount trend data by organization to Figure 3.
For example, in accordance with the binding information (header, display data, title) shown in FIG. 30(A), the sales amount transition data by organization shown in FIG. 30(B) is bound to Chart 3 as shown in FIG. 30(C).

In Figure 3, the title is "Forecast and Actual Results by Organization" (50th Period: April 2022 - March 2023), the row items are business establishment and department, and the column items are fiscal year and month, sales amount (forecast), and sales amount (actual).

4. Bind the sales amount trend data by construction to Figure 4.
For example, in accordance with the binding information (header, display data, title) shown in FIG. 31(A), the sales amount trend data by construction work shown in FIG. 31(B) is bound to Chart 4 as shown in FIG. 31(C).

In Figure 4, the title is "Forecast and Actual Results by Project" (Projects managed by Business A, 50th period: April 2022 to March 2023), the row items are the project name, and the column items are the fiscal year and month, sales amount (forecast), sales amount (actual), and forecast progress rate.

5. Output Charts 1 to 4 to the analysis screen Output bound Charts 1 to 4 to areas A2 to A5 on the analysis screen, respectively, as shown in FIG.

6. Highlight the data that is subject to an abnormality based on the construction name information contained in the "abnormality judgment result data" linked to the message.
Figure 33A (A) shows an example of graph coloring parameters, Figure 33A (B) shows an example of abnormality judgment result data, Figure 33A (C) shows an example of cumulative progress rate trend - actual data (data bound to graphs 1 and 2), Figure 33A (D) shows an example of cumulative progress rate trend - approximate curve data, and Figure 33B (E) shows an example of the display on the analysis screen.

In accordance with the graph coloring parameters, the plots, lines, and legends in Graphs 1 and 2 are highlighted in red. As shown in Figure 33A (A), the graph coloring parameters include the legend's abnormality flag (True or False), reference information flag (True or False), legend color, and line color. The reference information is used to confirm how abnormal the object being determined to be abnormal (construction work in this job) is.

When the name of the project in the legend is the same as the name of the project for which an abnormality was detected, the color corresponding to the True abnormality flag in the legend will be applied. For legends other than the project for which an abnormality was detected, the color corresponding to the False abnormality flag in the legend will be applied.

For both graphs, the normal data threshold is set as a dashed line. For graph 2, the past performance data does not display plots because the fiscal years and months of construction start dates vary.

The points to note about the initial display of the analysis screen are as follows:
(1) A message is displayed in area A1 of the analysis screen.
It outputs detailed information including the detected anomaly, the detection method, and the standard value used to determine the anomaly. For the user checking the information, analysis can be performed from a state where the construction work that may have an anomaly is known.
(2) Graphs 1 and 2 and Tables 3 and 4 are displayed in areas A2 to A5 of the analysis screen.
Graph 1 allows you to check the cumulative progress rate transition of an approximate construction project based on past construction work. It is possible to assume that future construction work will also basically follow this approximate cumulative progress rate transition. It is also possible to check how much the cumulative progress rate of a construction project for which an abnormality has been detected deviates from the predicted cumulative progress rate approximation curve. Construction projects for which an abnormality has been detected are considered to be construction projects where the following issues may occur, rather than "unfinished construction projects."
・High workload work towards the end of the construction period ・Delayed completion of construction work Check the increasing trend of the cumulative progress rate by looking at the change in the cumulative progress rate for each construction period progress rate. In this example, it can be seen that the cumulative progress rate of the construction work where an abnormality was detected is lower than that of other construction work.

Graph 2 shows the trend in the cumulative progress rate for construction work where an abnormality was detected, by fiscal year and month. You can see how there is a deviation from the cumulative progress rate approximation curve calculated from past construction work, by month. It is highly likely that work carried out in months with large deviations is the cause of the deviation. In this example, we can see that from fiscal year and month 2022/05 onwards, the actual results have exceeded the normal range from the approximate value (within the lower progress rate threshold).

Table 3 allows you to check the actual results of construction work by organization that occurred within a specified fiscal period. The actual results of the organization managing the construction work for which an abnormality was detected may deviate significantly from the predicted values. In addition, since you can check the sales results for construction work that only occurred within the fiscal period, it can be used to analyze the current period's sales results. Since the predicted values can be used as a reference for possible sales plans, they can be used to predict the actual sales results for the current period. Figure 34 is a diagram that explains how to check discrepancies in Table 3. In this diagram, for example, if the difference between the forecast and actual results as of August is 10,000 and the difference between the forecast and actual results as of November is 74,800, it can be seen that the discrepancy amount as of November has increased abnormally.

In Table 4, you can check the actual results for each project managed by the business office that manages the project where an abnormality was detected. If the cause of the abnormality in the project where an abnormality was detected lies within the organization, there is a possibility that an abnormality has also occurred in another project under management. Figure 35 is a diagram that explains how to check the discrepancy amount in Table 4. In the figure, for example, assume that an abnormality was detected in project A this time. It can be confirmed that the actual results for project E also deviate significantly from the predicted value.

(Points to check)
An example of analyzing the progress of the completed work (cumulative progress rate by construction progress rate) by switching to each construction type will be described. Figure 36 is a diagram for explaining an analysis image using Graph 1, where (A) shows the case of construction type, and (B) shows the case of demolition type.

In many cases, the progress rate trend will differ depending on the characteristics of the construction project, such as the type of work. The following types of construction projects may have similar progress rate trends:

・Construction period condition specification image: Divide into one-month units and calculate forecast values. For example, 6-month construction period, 7-month construction period, 8-month construction period, ... 24-month construction period (2-year construction period) ... etc. Divide into several-month units and calculate forecast values. For example, 6-12-month construction period (6-month span), 12-24 months (1-year span) ... etc.
- Image of specifying work type conditions: Explained in detail in this example.
・Condition specification image of public/private classification (public works or private works): Calculate forecast value for public works, calculate forecast value for private works, etc.

Few companies know in advance which characteristics to limit their analysis to in order to detect abnormalities, or whether these characteristics are appropriate for progress rate analysis, so in many cases analysis is required to identify these characteristics. By performing the analysis to identify these characteristics, explained in Processing Flow 2 (3-2) 1 and 2, feature by feature, you can analyze where your company's characteristics appear.

For example, if you want to calculate forecast values by dividing the construction period into one-month or several-month increments, you can do so by adding a row to the construction data acquisition range condition data, as shown in Figure 37, and registering, for example, the condition "Determination Range" and condition value "1." If this column is added, additional information is passed to the parameters used when acquiring data for anomaly detection, and data acquisition is performed.

(3) Switch the display of the analysis graph and perform the analysis (if no specific conditions are specified)
(3-1) When calculating the predicted value at the time of anomaly detection, an analysis is performed on cases where the characteristics of progress trends emerge when all construction work of the executing company is targeted without specifying specific conditions (construction period, type of work, etc.). The predicted value is calculated without specifying the "data reference unit" of the [construction data acquisition range condition data] used when determining anomalies. Essentially, the same processing is performed as in 1: Anomaly detection and 2: Initial screen display processing in this processing flow. The differences in processing are described below.

1. Anomaly detection 1-1. Obtain the range conditions for data to detect anomalies.
The set construction data acquisition range condition data and abnormality determination result registration target period data are acquired.
38 is a diagram showing an example of construction data acquisition range condition data. In this figure, no condition value is set. Here, a pattern in which no condition value is set will be explained (the progress rate transition is analyzed for all construction projects).

1-2. The construction data in the business database 106a is referenced to calculate the predicted progress rate for each construction project and construction period progress rate.
The data is acquired by setting the parameters to values within the acquisition range condition data for the construction data acquired in 1-1. Using the parameters (data reference unit unspecified) as shown in Figure 39(A) as a key, the construction data in the business database 106a is referenced and the construction data as shown in Figure 39(B) is acquired.

From the construction data obtained in 1-3.1-2, since the data reference unit is not specified this time, the data for all construction work will be used as the population, and the cumulative progress rate (approximate value) by construction period progress rate will be calculated based on this population.

Figure 40 (A) shows the acquired construction data, and Figure 40 (B) plots each construction project in the construction data, with the horizontal axis showing the construction progress rate (%) and the vertical axis showing the cumulative progress rate (%).

Figure 41 is a diagram for explaining the calculation of the approximate curve F(x), predicted progress rate, and upper and lower limits. As shown in Figure 41(A), the completed (past) construction work is used as the population, and the approximate curve F(x) of the cumulative progress rate by construction period progress rate for the population is calculated. In addition, the standard deviation function S(x) is calculated by focusing on the tendency of data dispersion within the population. Then, as shown in Figure 41(B), the construction period progress rate is substituted into the approximate curve F(x) and the standard deviation function S(x), respectively, to calculate the predicted progress rate and upper and lower limits for the uncompleted construction work. In this way, the predicted progress rate is calculated using F(x), and the upper and lower limits are calculated using S(x).

For example, the predicted value and upper and lower limits are calculated using the parameters shown in Figure 42A (A) (X-axis item: construction progress rate, Y-axis item: cumulative progress rate, approximate function candidates [polynomial function, logistic function, Bass model function], etc.). Figure 42A (B) shows an example of approximate data for the cumulative progress rate calculated by construction progress rate, and includes items for construction progress rate, predicted value (predicted value of cumulative progress rate), lower limit, and upper limit. Note that this explanation will use values that have already been scaled, so an explanation of scaling will be omitted. After calculating F(x), the data dispersion is calculated to calculate the standard deviation function S(x). The construction progress rate is substituted for x in F(x) and S(x), and the predicted value and upper and lower limits of F(x) are calculated. The same function is used to calculate the predicted value and upper and lower limits for each work type and project.

Figure 42B (D) shows the construction data after the predicted value and upper and lower limits have been calculated, and in addition to the predicted value (predicted progress rate), lower limit, and upper limit, information on the type of work is also stored.

According to the "acquisition criteria" for the anomaly determination result registration target period data acquired in 2-1, shown in Figure 42B (C), a determination is made as to whether there are any anomalies in the data for the month prior to the month to which the anomaly determination execution date belongs. In this case, the execution date is 2022/11/07, so a determination is made as to whether there are any anomalies on 2022/10.

2-4. Obtain data for graph display.
In step 2 of this process flow, the work type in parameters (2-3) to (2-5) is left blank (unspecified) to acquire data. Since there is no difference other than the addition of data extraction conditions, we will omit the explanation here.

2-5. Bind the graph display data to the graph, color the graph, and complete the display process. This is the same process as step 2 in this processing flow, so a detailed explanation will be omitted. Figure 43 shows the final image that will be drawn.

(Points to check)
It is assumed that it will be used in many cases when the specific conditions for capturing the trend of completed work are not known in advance. First, check the company-wide trend of completed work, and if the results are not useful, specify the characteristics that are likely to show a trend and check them again to grasp the trend of your own company. Depending on the company, there are cases where the possible completed work trend of the construction can be grasped once the company-wide trend of completed work is confirmed.

Figure 44 is a diagram to explain an example of checking company-wide trends. There are no plot characteristics such as slopes or concentrations, and the company's trends are not apparent, making it impossible to grasp trends in construction volume.

Figure 45 is a diagram explaining an example of extraction and confirmation using specific conditions. As shown in Figure 45(A), approximate cumulative progress values that can occur for each construction period can be confirmed. A trend that appears to be an upward trend in the volume of work being completed emerges. Figure 45(B) is a diagram showing the actual state of this job when it is used; there are cases where such a trend can be identified when checking across the entire company without narrowing down the specific conditions.

(3-2) Check and analyze the actual and forecast volumes that occurred in past accounting periods.
As shown in Figure 46A (A), you can specify the accounting period item in the extraction conditions to switch the data to be output for analysis. If the current period is set to 50 (2022/04-2023/03), switch to 49 (2021/04-2022/03) (the previous period).

46B(B) is a diagram showing an example of the display on the analysis screen after switching the accounting period. For graphs affected by the accounting period extraction items, the graph display is switched according to the specified value as follows.
・Graph 1: Only construction work that occurred during the specified period is output to the graph ・Graph 2: If construction work that was determined to be abnormal has been performed during the specified period, the graph is output; if not, the graph is not output ・Table 3: Switch the displayed information for the period and the previous period/construction period information to match the specified accounting period ・Table 4: Switch the displayed information for the period and the previous period/construction period information to match the specified accounting period

The following describes the process from data acquisition to screen display, but as the process is basically the same as in (2), we will only explain the parts where the process differs.

1. Cumulative progress rate transition - actual results data is obtained from the construction data in the business database 106a. Based on the conditions initially set in the extraction conditions, cumulative progress rate transition - actual results data is obtained and used as data for display in Figures 1 and 2.

The parameters shown in Figure 47A (A) do not specify the type of work. Based on the parameters (keys) of start period "2021/04," end period "2022/03," and work "Work A," the work data shown in Figure 47B (D) is extracted, and the cumulative progress rate trend - actual data shown in Figure 47B (C) is obtained.

In this case, if there is even one record that contains an "accounting year and month" within the period from the start to the end of the period, that construction project will be eligible for acquisition. For example, as shown in Figure 47B (B), in this example, construction projects A, B, and C will all be eligible for acquisition.

2. Obtain "sales amount trend data by organization" from the cumulative progress rate trend_actual data obtained in 1 and the abnormality determination result data in the abnormality determination result data table 106c.

Figure 48B (D) shows an example of the cumulative progress rate trend - actual data obtained in (2-3), Figure 48B (E) shows an example of anomaly determination result data, and Figure 48A (C) shows an example of sales amount trend data by organization.

Using the start period "2021/04" and end period "2022/03" as parameters (keys) as shown in Figure 48A (A), the cumulative progress rate trend - actual data shown in Figure 48B (D) and the abnormality determination result data shown in Figure 48B (E) are used to obtain the sales amount trend data by organization shown in Figure 48A (C) as data for display in Chart 3. Graph 3 checks only the construction results that occurred within the relevant accounting period, and obtains and displays only the data for the results within the box shown in Figure 48A (B).

In the example for the fiscal year and month "2022/03", the results would be: Department A: Actual results "200", forecast "600", Department B: Actual results "9,000", forecast "9,200", Business location A (Department A + Department B): Actual results "9,200", forecast "9,800".

3. The sales amount trend data by construction project is obtained as display data for Figure 4 from the cumulative progress rate trend - actual data obtained in 1 and the abnormality determination result data in the abnormality determination result data table 106c.

Figure 49A (C) shows an example of data on the trend in sales amount by construction project, Figure 49B (D) shows an example of data on the trend in cumulative progress rate - actual results data obtained in (2-3), and Figure 49B (E) shows an example of data on the abnormality determination results.

Using the start period "2021/04," end period "2022/03," and project "Project A" as parameters (keys) as shown in Figure 49A (A), the sales amount trend data by project shown in Figure 49A (C) is obtained as data for display in Figure 4 from the cumulative progress rate trend - actual data shown in Figure 49B (D) and the abnormality determination result data shown in Figure 49B (E). If there is even one record in the above period that includes the "fiscal year and month," that project will be acquired. For example, in the table of Figure 49A (B), projects A, B, and C will be acquired.

4. A graph of the acquired data from steps (2) and 1-3 is output to the analysis screen. Figure 50 shows an example of the analysis screen display.

(Points to check)
FIG. 51 is a diagram for explaining points to be checked focusing on Tables 3 and 4. FIG. 51(A) is a diagram for explaining points to be checked in Table 3. In the diagram, it is possible to check the actual sales performance of construction work that occurred within a specified accounting period. When construction work for which an abnormality has been detected spans multiple periods, it is possible to check the actual situation by checking the information for the period in which a large deviation from the predicted value occurs.

Figure 51 (B) is a diagram explaining the check points in Table 4. In this figure, it is possible to check the progress of construction work as of the specified accounting period. For construction work in which an abnormality has been detected, it is possible to analyze by period when the deviation from the predicted value began to widen.

[3-3. Calculation methods for approximate curves, predicted values, upper and lower limits]
The calculation method used by the detection unit 102b to calculate the approximate curve, forecast value, and upper and lower limit values will be described in detail with reference to Figures 52 to 75. While the above process targets construction data, the following example will be described using monthly actual data. While the target data is different, the calculation method is similar. The explanation of the calculation method here also applies to the above process. Figure 75 illustrates the correspondence between the items (item names used in the description of the present invention) used in the description of [3-1. Overall Processing] and [3-2. Sample Data], which are characteristic features of the present invention, and the items (item names used in the description of the approximate curve algorithm) used in the description of [3-3. Method for Calculating the Approximate Curve, Forecast Value, and Upper and Lower Limit Values]. The item names "Percentage of Progress,""CumulativeProgress," and "Actual Sales Amount" used in the description of the present invention correspond to the item names "Percentage of Elapsed Construction Period,""ActualProgress," and "Actual Amount," respectively, used in the description of the approximate curve algorithm.

Figure 52 shows an example of the configuration of the abnormality determination definition master 106d. The abnormality determination definition master 106d registers the abnormality determination definition ID, abnormality determination definition name, data acquisition definition, algorithm used, and parameter settings in association with each other.

In the example shown in the figure, the anomaly determination definition ID is "JD001", the anomaly determination definition name is "Project progress alert", the algorithm used is "Approximation curve", and the parameter settings are {X-axis item: number of months elapsed, Y-axis item: cumulative actual amount, X-axis scale unified item: planned construction period, Y-axis scale unified item: actual amount, approximate function candidate name: [polynomial function, logistic function, Bass model function], smoothness: 12, significance level: 0.1}.

(T1: Approximate curve calculation process)
The details of the approximate curve calculation process will be explained with reference to Figures 53A to 55. In the approximate curve calculation process, the data is not distinguished by case code, but an approximate curve is calculated using approximate components based on all case data of past case monthly performance data. This process has the following features.

(1) The options for approximation functions include polynomial functions, exponential functions, logarithmic functions, logistic functions, and Bass model functions, and multiple options can be selected.
(2) Among the selected functions, the one with the best AIC (Akaike Information Criterion), an index for measuring the accuracy of approximation, will be the approximate curve used in the subsequent processing flow.

Hereafter, the approximation function deemed to have the best approximation accuracy will be referred to as "Y = F(X)", the name of this approximation function will be called the "performance-centered approximation function", and the curve that results when this approximation function is plotted on a graph will be called the "performance-centered approximation curve". X represents the planned completion rate, Y represents the actual progress rate, and F represents the performance-centered approximation function. By passing X (the planned completion rate) to F, the predicted value of Y (the actual progress rate) can be calculated.

The above "bus model function" is used by the Ministry of Land, Infrastructure, Transport and Tourism to estimate the progress rate of construction work. In addition, polynomial functions and logistic functions are also used for comparison with the bus model function (https://www.mlit.go.jp/sogoseisaku/jouhouka/content/001348995.pdf).

Specifically, the following processing is performed:

(1) Among the parameter settings registered in the abnormality determination definition master 106d, information on the X-axis item, Y-axis item, and function for each approximate function candidate name is passed to the approximating component to find the optimal coefficient.
(2) Calculate the AIC based on the coefficients returned from the approximate parts.
(3) The approximate function with the smallest AIC is set as the performance center approximate function.

Each process will be explained in detail below.
(1) Among the parameter settings registered in the abnormality determination definition master 106d, information on the X-axis item, Y-axis item, and function for each approximate function candidate name is passed to the approximating component to find the optimal coefficient.

In Figures 53A and 53B, (A) shows an example of data from the anomaly determination definition master 106d. In this example, the anomaly determination definition ID is "JD001," the anomaly determination definition name is "Project progress alert," the algorithm used is "Approximation curve," and the parameter settings are {X-axis item: number of months elapsed, Y-axis item: cumulative actual amount, approximate function candidate names: [polynomial function, logistic function, Bass model function], ...}. The options for the approximation function correspond to the approximate function candidates in the parameter settings for the data registered in the anomaly determination definition master 106d, and the X and Y values of the data to draw an approximate curve correspond to the processed values in the X value item and Y value item in the parameter settings for the data registered in the anomaly determination definition master 106d.

(B) shows an example of monthly performance data for a past project after processing. (C) shows the approximate function candidates, including polynomial functions, logistic functions, and bus model functions. If a polynomial function is selected, polynomial function expressions of degrees 0 to 6 will be passed. In this example, polynomial functions, logistic functions, and bus model functions have been selected as approximate function candidates, so the nine functions to be approximated that will be passed to the approximation component are the ones shown above. Exponential functions and logarithmic functions may also be selected as approximate function candidates.

The following process is performed for each candidate approximation function. As an example, we will explain the case where the function to be approximated is y = a/(1 + ebx + c) (logistic function). (D) shows an example of information (1) (data to be passed as an approximate curve) passed to the approximation component. (E) shows an example of information (2) (function to be approximated) passed to the approximation component. (F) shows an example of the information (coefficients) returned from the approximation component, where a = 100.85, b = -0.069, and c = 3.68.

(2) Calculate the AIC based on the coefficients returned from the approximate parts. For information on calculating the AIC, see http://www.radio3.ee.uec.ac.jp/ronbun/TR_YK_048_AIC.pdf.

In Figure 54, (A) shows the formula for calculating AIC, which is AIC = (data size) x log (sum of squares of differences from predicted value) / (data size) + 2 x (number of coefficients). Here, the predicted value of Y is calculated by substituting X, the data for which you want to draw an approximate curve, into the approximation function. The square of the difference from the predicted value of Y is calculated as the square of (Y value - predicted value of Y) of the data for which you want to draw an approximate curve. The data size is the number of data (X, Y) pairs.

A characteristic of AIC is that the smaller the sum of squares of the difference from the predicted value, the smaller the AIC; and the larger the number of coefficients, the larger the AIC. The AIC value itself has no meaning; comparing AIC values is an indicator used to determine the superiority of an approximation function. The AIC formula is explained below. The first term in the AIC formula, "(data size) x log (sum of squares of the difference from the predicted value) / (data size)," represents the accuracy of the data for which the approximate curve is to be drawn, while the second term, "2 x (number of coefficients)," represents the penalty for having a large number of coefficients. By determining the superiority of an approximation function based on AIC, a well-balanced approximation function that is robust to data changes and does not overly fit the data for which the approximate curve is to be drawn is selected. The benefits of determining the superiority of an approximation function based on AIC are explained below. Even if the amount of actual data from past projects increases over time, an approximation function that overly fits past data is not selected. This provides the following two benefits: (1) It prevents the type and order of the approximation function from frequently changing each time an anomaly is detected. (2) Even if the number of projects that deviate significantly from the planned progress is dramatically reduced due to thorough project progress management, the approximation curve will shift gradually before and after thorough management.

(B) shows the data (X, Y) for which you want to draw an approximate curve. (C) shows the predicted value of Y and the sum of squares of the difference between the predicted value of Y and the data.

When the data size, the sum of squares of the difference between the predicted value of Y, the coefficient, and the number of coefficients (3) are substituted into the AIC calculation formula, the AIC becomes 94.18, as shown in (D).

When the above process is completed for all approximate function candidates, data such as that shown in Figure 55(A) can be obtained, and the AICs for all approximate function candidates can be obtained. In this figure, actual numerical values are entered in the "..." boxes.

(3) The approximate function with the smallest AIC is set as the performance center approximate function.
The explanation of the processing flow will proceed under the assumption that, after calculating the AIC for all approximate function candidates, the AIC for the logistic function was the smallest. If the performance-centered approximation function is Y = F(X), in this explanation, the performance-centered approximation function will be y = 100.85/(1 + e ^{- 0.069X + 3.68} ), as shown in Figure 55(B). This is the function with the smallest AIC, and the coefficients are set to the function to be approximated. Hereinafter, this function will be referred to as F(X).

(T2: Processing to calculate the dispersion of data from the approximation curve)
The process of calculating the dispersion of data from the performance center approximation curve will be described in detail with reference to Figures 56 and 57. The process of calculating the dispersion of data from the performance center approximation curve involves the following steps.

(1) Calculate the predicted value of the actual progress rate for each scheduled progress rate of the monthly performance data of past projects.
(2) Calculate the difference (residual) between the predicted value and the actual data of the performance progress rate.
(3) Calculate the square of the residual (square of the residual).

A specific calculation will be described with reference to FIG.
(1) The predicted value of the actual progress rate is calculated as follows: predicted value = F (progress rate of construction period).
(2) The residual, which is the difference between the actual progress rate and the predicted value, is calculated as follows: Residual = Actual progress rate - Predicted value.
(3) Calculate the square of the residual: residual squared = (residual) ² .

Figure 56 (A) shows the planned completion time lapse rate [%] and actual progress rate [%] of actual data for past projects. Figure 56 (B) shows the newly added columns: forecast value [%], residual [%], and squared residual [% ² ]. The forecast value [%] is calculated by passing the planned completion time lapse rate to the actual performance center approximation function F(X). Because the residual is squared, all residual squares are greater than or equal to 0. By squaring the residual, the magnitude of the squared residual can be expressed as the degree to which the past performance data deviates from the approximation curve.

Figure 57 (A) is a graph showing predicted values and approximate curves for past project data. The horizontal axis shows the planned progress rate [%], and the vertical axis shows the actual progress rate [%]. Past actual data is plotted, and predicted values are plotted on the approximate curve. The actual progress rate minus the predicted value shows the residual.

Figure 57 (B) is a graph showing the residuals, with the horizontal axis representing the progress rate [%] and the vertical axis representing the residuals [%].

FIG. 57(C) is a graph showing the square of the residual, with the horizontal axis representing the progress rate [%] and the vertical axis representing the square of the residual [% ² ].

(T3: Processing to calculate the tendency of data dispersion)
The process of calculating the tendency of data dispersion will be described in detail with reference to Figures 58A to 60. The process of calculating the tendency of data dispersion involves the following steps.

(1) Draw an approximation curve for the square of the residual, name this approximation function the residual square approximation function, and let this residual square approximation function be σ ² = V(X). The residual square approximation function is in the form of (polynomial function) ^2. By passing X (the planned completion time progress rate) to the residual square approximation function, the predicted value of the residual square at that planned completion time progress rate will be returned.
(2) Take the square root of the residual square approximation function, name this function the standard deviation function, and let this standard deviation function be σ = S(X). By passing X: the planned completion period progress rate to the standard deviation function, the standard deviation (the degree of dispersion of the actual progress rate) for that planned completion period progress rate is returned.

By taking the square root of the residual squared approximation function, we can create a function that returns the standard deviation. The specific process is explained below.

(1) Draw an approximate curve using approximate parts for the square of the residual (this curve is called the residual square approximate curve).
58A and 58B, (A) shows the planned completion period elapsed rate [%], actual progress rate [%], forecast value [%], residual [%], and square of residual [% ² ] of the actual data of past projects. (B) shows candidates for the residual square approximation function.

(Polynomial Function) The reason why function ² is specified as the function to be approximated is as follows.
1. Because we are not approximating the progress rate of the project, there is no need to specify special functions such as bus model functions that are said to represent the progress of the project, so we specify as simple a function as possible.
2. When approximating the actual progress rate, the actual progress rate increases as the progress rate of the scheduled completion period progresses, but the squared residual may increase or decrease as the progress rate of the scheduled completion period progresses. Therefore, (polynomial function) ² is used to capture the increase or decrease in the squared residual.
3. (Polynomial function) By setting it to ² , the square root of the residual square approximation function (polynomial function) can be used later as the standard deviation function.

The following process is performed for each candidate residual squared approximation function. As an example, we will explain the case where the function to be approximated is y = ( ^ax² + bx + c) ^² . In Figures 58A and 58B, (C) shows information (1) (data for which an approximate curve is to be drawn) passed to the approximating part, and (D) shows information (2) (function to be approximated) passed to the approximating part. (E) shows the information returned from the approximating part, with coefficients a = -0.0028, b = 0.3209, and c = 2.1336.

Next, the AIC is calculated. This is the same process as (T1: Approximate curve calculation process) in the processing flow, so a detailed explanation will be omitted.

When the above process has been completed for all candidate functions that approximate the square of the residual, the data shown in Figure 59 (A) will be obtained.

In this example, the function with the smallest AIC, with coefficients set to the function to be approximated, is the residual square approximation function y = (-0.0028x ² + 0.3209x + 2.1336) ² , as shown in Figure 59 (B). Hereafter, the following explanation of the processing flow will proceed under the assumption that the AIC of this (quadratic function) ² function was the smallest. The residual square approximation function is assumed to be σ ² = V(X). In this example, V(X) = (-0.0028X ² + 0.3209X + 2.1336) ² .

59(C) shows a graph in which the residual square approximation curve V(X) is plotted against the square of the residual, with the horizontal axis representing the planned completion period progress rate [%] and the vertical axis representing the residual square [% ² ]. By drawing an approximation curve in this way, it is possible to take into account the variation in data according to the completion period progress rate, such as small deviations from the approximation curve in the early and late stages from the start of the project and large deviations in the middle stages.

(2) Calculate the standard deviation function.
When the planned completion progress rate is passed to X of the residual square approximation function calculated in (1), the output value is in units of [% ² ], so it is converted to a standard deviation function in units of [%] by taking the square root. The standard deviation function S(X) is calculated as S(X) = √V(X). In this processing flow example, V(X) = (-0.0028X ² + 0.3209X + 2.1336) ² , so S(X) = -0.0028X ² + 0.3209X + 2.1336.

Figure 60(A) shows the progress rate, progress rate, forecast value, residual, and the square of the residual plus the absolute value of the residual for the performance data of past projects. Figure 60(B) shows a graph displaying the standard deviation function V(X) as a standard deviation curve, with the horizontal axis representing the progress rate [%] and the vertical axis representing the absolute value of the residual [% ² ]. The absolute value of the residual represents the distance of deviation between the performance data of past projects and the performance center approximation curve. Note that in actual calculation processing, the absolute value of the residual is not calculated from the residual between the performance data of past projects and the forecast value using the performance center approximation function, but it is provided as a reference when displaying the standard deviation curve on a graph.

(T4: Calculation process of predicted values and thresholds for cases subject to abnormality detection)
The details of the calculation process of the predicted value and threshold for an abnormality detection target case will be explained with reference to Figures 61 to 64. The actual result center approximation function F(X) and standard deviation function S(X) for finding the predicted value of the actual progress rate have been prepared through the processing up to this point. In the calculation process of the predicted value and threshold for an abnormality detection target case, past case data is not used, but the actual data of the abnormality detection target case is used for the calculation. In the calculation process of the predicted value and threshold for an abnormality detection target case, the following processing is performed on the processed actual data of the abnormality detection target case.

(1) A predicted value is calculated using the planned completion period progress rate/actual result center approximation function F(X).
(2) Calculate the standard deviation (the predicted value of the distance of deviation from the approximate curve) using the scheduled construction period progress rate/standard deviation function S(X).
(3) The upper and lower thresholds are set by providing a range above and below the predicted value equal to the standard deviation multiplied by a constant value calculated based on the significance level.

(1) Calculate the predicted value. Calculate the predicted value of the actual progress rate for the project to be determined as abnormal based on the planned progress rate.

Figure 61 (A) shows the processed actual data for a project subject to anomaly detection, with columns for predicted values added to the project code, fiscal year and month, progress rate, and actual progress rate. The predicted values are calculated using the performance center approximation function F(X), where x = progress rate.

Figure 61 (B) is a graph of the predicted values in Figure 61 (A), with the horizontal axis representing the planned progress rate [%] and the vertical axis representing the actual progress rate [%].

(2) Calculate the standard deviation. Calculate the standard deviation of the planned completion period progress rate for projects subject to abnormality detection after processing.

Figure 62 (A) shows the processed performance data for a project subject to anomaly detection, with the standard deviation column added to the project code, fiscal year and month, planned completion period progress rate, actual progress rate, and forecast value. The standard deviation is calculated using the standard deviation function S(X) where x = planned completion period progress rate. The calculated standard deviation means that there is a 68% chance that the performance data will fall within a range that is the standard deviation away from the performance center approximation curve.

FIG. 62(B) is a graph of the standard deviation of FIG. 62(A), with the horizontal axis representing the progress rate [%] and the vertical axis representing the standard deviation [% ² ].

(3) Determine the threshold value.
Figure 63 (A) shows the processed monthly performance data for a project to be judged as abnormal, with columns for upper threshold [%] and lower threshold [%] added to the project code, fiscal year and month, planned completion period progress rate, actual progress rate, forecast value, and standard deviation.

As shown in Figure 63 (B), the thresholds are calculated as follows: upper threshold = predicted value + _Zα/2 × (standard deviation), lower threshold = predicted value - _Zα/2 × (standard deviation). _Zα/2 means the value of the upper α/2 percentile of the standard normal distribution. For this α, the value of the parameter setting (significance level) registered in the anomaly determination definition master 106d in premise (1) is used.

Figure 64 (A) shows an example of data in the abnormality determination definition master 106d, in which the parameter {significance level: 0.1} is set for the abnormality determination definition ID "JD001."

For example, if the significance level α is 0.1, then α/2 = 0.05 (5%), and Z _α/2 = 1.645. In this case, in Figure 64 (B), for the record of the abnormality judgment target project B001 where the planned completion period elapsed rate is 40%, the upper threshold = 28.678 + 1.645 x 10.415, and the lower threshold = 28.678 - 1.645 x 10.415. This can be interpreted as meaning that there is about a 90% probability that the actual performance progress rate will be above the lower threshold and below the upper threshold. Conversely, if this range is exceeded, it is judged to be abnormal.

Figure 64 (C) is a graph of the forecast values, upper threshold, and lower threshold of Figure 63 (B), with the horizontal axis representing the planned completion period elapsed rate [%] and the vertical axis representing the actual progress rate [%]. Note that while graphs of the actual performance center approximation curve and the upper and lower thresholds are shown, they are for reference purposes only.

Figure 64(D) is a diagram explaining the significance level and percentile (https://ai-trend.jp/basic-study/normal-distribution/normal-distribution/).

(T5: Abnormality determination process)
The abnormality determination process will be described in detail with reference to Figures 65 and 66. The abnormality determination process includes the following steps.

(1) Anomaly detection is performed based on the monthly performance data of the case to be detected and the threshold value obtained in the previous process.
(2) The data of the determination result is stored in a table (not shown) in the storage unit 206 .

The abnormality determination process will be specifically described below.
(1) For a case to be judged as an anomaly, an anomaly is judged by comparing the actual progress rate of the case to be judged as an anomaly with the threshold value obtained in the previous process.

Figure 65 (A) shows the project code, fiscal year and month, progress rate, actual progress rate, forecast value, standard deviation, upper threshold, and lower threshold for the actual data of the project subject to abnormality detection.

Figure 65 (B) shows a graph of the abnormality determination in Figure 65 (A), with the horizontal axis representing the planned completion period elapsed rate [%] and the vertical axis representing the actual progress rate [%], and plotting the actual performance center approximation curve, upper threshold, lower threshold, B001, and B002. In the figure, B001's planned completion period elapsed rate of "40" and actual progress rate of "60" are determined to be abnormal because they exceed the upper threshold. Also, B002's planned completion period elapsed rate of "57.143" and actual progress rate of "20" are determined to be abnormal because they fall below the lower-upper threshold. For example, if the upper threshold is exceeded, the possibility of fraudulent overstating of actual performance is detected. Also, if the lower threshold is below the lower threshold, the possibility of a delay in progress is detected.

(2) The judgment result data is stored in a judgment result table.
66 is a diagram showing an example of judgment result data. The judgment result data may include the case code, accounting year and month, progress rate for the planned completion period, actual progress rate, forecast value, standard deviation, upper threshold, lower threshold, and judgment result (FALSE or TRUE). For data judged to be abnormal, the judgment result is set to "TRUE", and for other data, it is set to "FALSE".

(T6: Calculation process of display data)
The display data calculation process will be explained in detail with reference to Figures 67 to 69. Through the process up to this point, the performance center approximation function F(X) and standard deviation function S(X) based on past cases have been determined. In the display data calculation process, data for display (to display the approximation curve on the screen) is created and the results are stored in a table for auxiliary information.

The display data calculation process involves the following steps.
(1) The progress rate of the construction period is calculated to display the data for display.
(2) The values on the approximation curve of the data to be displayed, and the values of the upper and lower thresholds are calculated.
(3) The results of the approximate curve algorithm are stored in the attached information table.

The specific processing contents will be explained below.
(1) The progress rate of the construction period to be displayed as data for display is calculated.
The progress rate of the construction period of the displayed data is obtained by dividing the minimum and maximum values of the progress rate of the construction period of the past project data by the value of a parameter called "smoothness." Figure 67 shows a graph of the actual data of past projects, with the horizontal axis representing the planned progress rate [%] and the vertical axis representing the actual progress rate [%], and A001, A002, and A003 being plotted.

This range of 0-120% of the planned progress rate is divided into 12 equal parts with a smoothness of 12 to prepare the progress rate of the progress rate for display data. For the smoothness, the parameter setting (smoothness) value registered in the anomaly judgment definition master 106d in premise (1) is used.

Figure 67 (B) shows an example of data settings for the abnormality determination definition master 106d. In the example shown in the figure, the parameter "Smoothness: 12" is set for the abnormality determination definition ID "JD001."

When the display data is displayed on the screen, the predicted value, upper threshold, and lower threshold may be displayed as a line graph, so the larger the "smoothness" value, the smoother the line graph will be, closer to a curve.

By dividing 0-120% into 12 equal parts, the construction progress rate for display data as shown in Figure 67 (C) is prepared. In this example, the interval between each construction progress rate is 10%, which is 12 equal parts of 0-120%, and the number of points for display data is 13.

(2) The values on the approximation curve of the data to be displayed, and the values of the upper and lower thresholds are calculated.
Using F(X) and S(X) obtained when processing the actual data of past cases, the values on the approximation curve for the progress rate of the construction period of the data to be displayed and the values of the upper and lower thresholds are calculated. Since this is exactly the same processing as that used to calculate the predicted values and thresholds for cases subject to abnormality detection, a detailed explanation of the calculation process will be omitted.

After the calculation process, the display data will be as shown in Figure 68(A), with columns for predicted value, upper threshold, and lower threshold added to the planned completion period elapsed rate. Figure 68(B) is a graph (line graph) of the display data in Figure 68(A), with the horizontal axis representing the planned completion period elapsed rate [%] and the vertical axis representing the actual progress rate [%], with the predicted value, upper threshold, and lower threshold plotted. The values of the white dots on the graph are calculated.

As can be seen from the graph, the width between the upper and lower thresholds and the performance center approximation curve is calculated using the standard deviation function, resulting in thresholds that reflect the following two characteristics:

1. In the early and later stages of a project, there is little chance of deviation from the approximate curve. Therefore, even a slight deviation from the approximate curve is highly abnormal, so the range of the threshold is narrow.
2. There is a high possibility of deviation from the approximate curve in the middle stage of a project. As a result, even if something is normal, there is a possibility of deviation from the approximate curve to some extent, and even if there is a slight deviation, it is not considered abnormal, so the range of the threshold is wide.

(3) The approximate curve result is stored in the auxiliary information table.
The following display data is stored in the approximate curve algorithm result attached information table (hereinafter referred to as the "approximate curve result attached information table"). The approximate curve result attached information table stores data to be displayed as supplementary information when the anomaly determination results are displayed as a graph on the screen. Unlike the data of the anomaly determination target case itself, data to be compared with the data of the anomaly determination target case is stored. The stored data is calculated based only on data from past cases.

Figure 69 (A) shows an example of display data, and Figure 69 (B) shows an example of approximate curve result ancillary information stored in the approximate curve result ancillary information table 106c based on the display data. The approximate curve result ancillary information includes the following fields: execution history ID, line number, X, Y, upper threshold, and lower threshold.

Here, the execution history ID is an ID automatically assigned for each execution of an abnormality determination, the row number is a consecutive number assigned to each row of display data starting from 0, X corresponds to the X-axis value of the display data, Y corresponds to the Y-axis value of the display data, the upper threshold is the upper threshold of the display data, and the lower threshold is the lower threshold of the display data.

If an abnormality is detected using the judgment result data (see Figure 66) and the approximate curve result ancillary information (see Figure 69 (B)), a message indicating the abnormality is detected and a graph (e.g., a line graph) of the abnormality judgment results is displayed.

(Options for processing past cases)
70 to 74, options for processing monthly performance data of past cases will be described. As described above, the anomaly determination definition master 106d is capable of setting scale unification items.

Figure 70 is a diagram showing an example of the settings for the anomaly determination definition master 106d. The assumed item set for the X-axis scale unified item is the planned construction period or implementation period, and the assumed item set for the Y-axis scale unified item is the budget or actual amount. The example of the settings for the anomaly determination definition master 106d shown in Figure 70 is an example for the construction industry, but it is also expected to be used in other industries.

The X-axis item, Y-axis item, X-axis scale unification item, and Y-axis scale unification item can be freely set according to the industry of the user of this anomaly detection system.

Figure 71 (A) shows an example of monthly performance data for past projects. Figure 71 (B) shows an example of planned and performance data for past projects.

Referring to Figure 72, four methods for processing data from past cases are explained. Data is classified into four patterns, A, B, C, and D, based on the parameter settings of the anomaly determination definition master 106d.

A. When the X-axis scale is unified: implementation period, and the Y-axis scale is unified: actual amount B. When the X-axis scale is unified: planned construction period, and the Y-axis scale is unified: actual amount C. When the X-axis scale is unified: implementation period, and the Y-axis scale is unified: budget D. When the X-axis scale is unified: planned construction period, and the Y-axis scale is unified: budget

Pattern names for patterns A, B, C, and D will be as follows: Pattern A will be "strict scheduled construction period and budget type", Pattern B will be "flexible scheduled construction period and strict budget type", Pattern C will be "strict scheduled construction period and flexible budget type", and Pattern D will be "flexible scheduled construction period and budget type".

"Strict" and "Flexible" mean strict and flexible checks, respectively, at the time of project completion to ensure that the work was completed within the planned timeframe and budget. An example of Pattern B is explained below.

(Pattern B. Flexible construction schedule and strict budget (X-axis scale unified item: planned construction schedule, Y-axis scale unified item: actual amount))
Pattern B, flexible scheduled construction period/strict budget type, will be explained with reference to Figures 73 and 74. An example of processing A003 will be explained below. Figure 73(A) shows an example of monthly actual data for A003, and Figure 73(B) shows an example of planned and actual data for A003. When the monthly actual data for A003 is processed and the planned construction period elapsed rate is calculated by dividing the number of months elapsed by the scheduled construction period, and the actual progress rate is calculated by dividing the cumulative actual amount by the actual amount, the result is as shown in Figure 73(C).

When the monthly performance data for all past cases is processed, it will look like Figure 74(A). Figure 74(B) is a graph of the monthly performance data for past cases in Figure 39(A), and Figure 74(C) shows the graph after calculation processing of the display data.

Since the maximum X-axis value for past projects is 120%, the threshold calculated in subsequent processing will converge to an X value of around 120%. This means that even if a project to be judged as abnormal is completed with a slight delay, it can be determined to be within the normal range. Furthermore, since the maximum Y-axis value for past projects is all 100%, the threshold calculated in subsequent processing will converge to a Y value of around 100%. This makes it possible to strictly check that projects to be judged as abnormal are completed according to the budget before they began.

As described above, this embodiment includes a detection unit 102b that divides the population of past case data according to specified specific conditions, calculates an approximate curve for the progress rate of the case data and the case progress rate for each population, calculates a predicted value for the progress rate by substituting the progress rate into the approximate curve function, calculates upper and lower limit values based on the approximate curve and the predicted value for the case progress rate, and compares the case data to be detected with the upper and lower limit values to detect anomalies, and a display control unit 102c that displays analysis data on the analysis screen for case data in which an anomaly has been detected, thereby enabling automatic detection of anomalies in case data and early detection and handling.

[4. Contribution to the United Nations-led Sustainable Development Goals (SDGs)]
This embodiment can contribute to improving business efficiency and promoting appropriate management decisions by companies, thereby contributing to the achievement of SDGs Goals 8 and 9.

Furthermore, this embodiment can contribute to reducing waste and promoting paperless and electronic systems, thereby contributing to SDGs goals 12, 13, and 15.

Furthermore, this embodiment can contribute to strengthening control and governance, thereby contributing to Goal 16 of the SDGs.

5. Other Embodiments
The present invention may be implemented in various different embodiments other than those described above within the scope of the technical concept set forth in the claims.

For example, among the processes described in the embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically using known methods.

In addition, the processing procedures, control procedures, specific names, registered data for each process, information including parameters such as search conditions, screen examples, and database configurations shown in this specification and drawings may be changed as desired unless otherwise specified.

Furthermore, with regard to the anomaly detection system 100, the components shown in the figure are functional concepts and do not necessarily have to be physically configured as shown.

For example, all or any part of the processing functions of the anomaly detection system 100, particularly the processing functions performed by the control unit, may be implemented by a CPU and a program interpreted and executed by the CPU, or may be implemented as hardware using wired logic. The program is recorded on a non-transitory, computer-readable recording medium containing programmed instructions for causing the information processing device to execute the processing described in this embodiment, and is mechanically read by the anomaly detection system 100 as needed. That is, a computer program that works in conjunction with the OS to issue commands to the CPU and perform various processes is recorded in a storage unit such as a ROM or HDD (Hard Disk Drive). This computer program is executed by being loaded into RAM, and works in conjunction with the CPU to form the control unit.

In addition, this computer program may be stored on an application program server connected to the anomaly detection system 100 via any network, and all or part of it may be downloaded as needed.

In addition, the program for executing the processing described in this embodiment may be stored on a non-temporary computer-readable recording medium, or may be configured as a program product. Here, this "recording medium" includes memory cards, USB (Universal Serial Bus) memories, SD (Secure Digital) cards, flexible disks, magneto-optical disks, ROMs, EPROMs (Erasable Programmable Read Only Memory), EEPROMs (registered trademark) (Electrically Erasable and Programmable Read Only Memory), CD-ROMs (Compact Disk Read Only Memory), MOs (Magneto-Optical disks), DVDs (Digital Versatile Disks), and more. This includes any "portable physical media" such as a Blu-ray Disc (registered trademark) and a Blu-ray (registered trademark) Disc.

A "program" is a data processing method written in any language or description method, and may be in any format, such as source code or binary code. It should be noted that a "program" is not necessarily limited to a single structure, but also includes a distributed structure consisting of multiple modules or libraries, or a structure that achieves its function by working in conjunction with a separate program, such as an OS. The specific structure and reading procedure for reading a recording medium in each device shown in the embodiments, as well as the installation procedure after reading, can use well-known structures and procedures.

The various databases stored in the memory unit are storage devices such as memory devices such as RAM and ROM, fixed disk devices such as hard disks, flexible disks, and optical disks, and store various programs, tables, databases, and web page files used for various processes and website provision.

The anomaly detection system 100 may be configured as an information processing device such as a known personal computer or workstation, or as an information processing device connected to any peripheral device. The anomaly detection system 100 may also be implemented by installing software (including programs, data, etc.) that causes the device to perform the processing described in this embodiment.

Furthermore, the specific form of distribution and integration of the devices is not limited to that shown in the figures, and all or part of them can be functionally or physically distributed and integrated in any unit depending on various additions or functional loads. In other words, the above-mentioned embodiments may be implemented in any combination, or embodiments may be implemented selectively.

100 Anomaly detection system 102 Control unit 102a Storage control unit 102b Detection unit 102c Display control unit 104 Communication interface unit 106 Storage unit 106a Business database 106b Anomaly detection execution data table 106c Anomaly determination result data table 106d Anomaly determination definition master 108 Input/output interface unit 112 Input device 114 Output device 300 Network

Claims (8)

An anomaly detection system including a control unit,
The control unit
configured to be able to access project data including one or more specific conditions, progress rates, and project progress rates;
a detection means for dividing a population of past case data according to specified specific conditions, calculating an approximate curve for the progress rate of the case data and the case progress rate for each population, calculating a predicted value of the case progress rate by substituting the progress rate into a function of the approximate curve, calculating upper and lower limit values based on the approximate curve and the predicted value of the case progress rate, and detecting an abnormality by comparing the case data to be detected with the upper and lower limit values;
a display control means for displaying analysis data on an analysis screen for the case data in which an abnormality has been detected ;
Equipped with
The project data includes construction data, the specific conditions include the type of work, the construction period, or the public/private classification, the progress rate includes the construction period progress rate calculated by the construction elapsed month/planned construction period, and the project progress rate includes the cumulative progress rate calculated by the currently incurred costs/(budgetary cost+currently incurred costs),
The anomaly detection system is characterized in that the construction data includes the construction, organization, one or more specific conditions, accounting year and month, sales amount, construction progress rate, and cumulative progress rate . The anomaly detection system described in claim 1, characterized in that the display control means displays a message in a specified area of the analysis screen based on the construction data, including the construction work detected as an anomaly, the detection method, and the cumulative progress rate determined to be an anomaly. The anomaly detection system described in claim 1, characterized in that the display control means displays, in a specified area of the analysis screen based on the construction data, the actual cumulative progress rate of past construction projects by population or of all past construction projects, by construction period progress rate, and its approximate curve, as well as a graph showing the trend in the actual cumulative progress rate linked to the construction project in which an abnormality was detected. The anomaly detection system described in claim 1, characterized in that the display control means displays, in a specified area of the analysis screen, the actual cumulative progress rate of past construction work and its approximate curve, by fiscal year and month, and a graph showing the trend in the actual cumulative progress rate linked to the construction work in which an abnormality has been detected, based on the construction work data . The anomaly detection system described in claim 1, characterized in that the display control means displays a table showing trends in sales performance and sales forecasts by organization within a specified accounting period in a specified area of the analysis screen based on the construction data. The display control means displays, in a predetermined area of the analysis screen based on the construction data,
The anomaly detection system according to claim 1 , characterized in that it displays a table showing the trends in sales performance and sales forecasts for each construction project by the organization in charge of the construction project in which the anomaly occurred. An anomaly detection method executed by an information processing device including a control unit,
The control unit
configured to be able to access project data including one or more specific conditions, progress rates, and project progress rates;
The control unit executes
a detection process of dividing a population of past business data according to specified specific conditions, calculating an approximate curve for the progress rate and progress rate of the project data for each population, calculating a predicted value of the project progress rate by substituting the progress rate into a function of the approximate curve, calculating upper and lower limit values based on the approximate curve and the predicted value of the project progress rate, and detecting anomalies by comparing the project data to be detected with the upper and lower limit values;
a display control step of displaying analysis data on an analysis screen for the case data in which an abnormality has been detected;
Including,
The project data includes construction data, the specific conditions include the type of work, the construction period, or the public/private classification, the progress rate includes the construction period progress rate calculated by the construction elapsed month/planned construction period, and the project progress rate includes the cumulative progress rate calculated by the currently incurred costs/(budgetary cost+currently incurred costs),
The anomaly detection method is characterized in that the construction data includes the construction, organization, one or more specific conditions, accounting year and month, sales amount, construction progress rate, and cumulative progress rate . An abnormality detection program to be executed by an information processing device having a control unit,
The control unit
configured to be able to access project data including one or more specific conditions, progress rates, and project progress rates;
The control unit
a detection process of dividing a population of past case data according to specified specific conditions, calculating an approximate curve for the progress rate of the case data and the case progress rate for each population, calculating a predicted value of the case progress rate by substituting the progress rate into a function of the approximate curve, calculating upper and lower limits based on the approximate curve and the predicted value of the case progress rate, and detecting anomalies by comparing the case data to be detected with the upper and lower limits;
a display control step of displaying analysis data on an analysis screen for the case data in which an abnormality has been detected;
This is an anomaly detection program for executing
The project data includes construction data, the specific conditions include the type of work, the construction period, or the public/private classification, the progress rate includes the construction period progress rate calculated by the construction elapsed month/planned construction period, and the project progress rate includes the cumulative progress rate calculated by the currently incurred costs/(budgetary cost+currently incurred costs),
The anomaly detection program is characterized in that the construction data includes the construction, organization, one or more specific conditions, accounting year and month, sales amount, construction progress rate, and cumulative progress rate.