Systems and/or methods for prediction and/or root cause analysis of events based on business activity monitoring related data

• US 8,140,454 B2
• Filed: 12/28/2007
• Issued: 03/20/2012
• Est. Priority Date: 12/28/2007
• Status: Active Grant

• Current Assignee: Software AG
• Original Assignee: Software AG
• Inventors: Bullen, Geoff, Dickson, Richard M., Roberts, Warren C.
• Primary Examiner(s): Gaffin, Jeffrey A
• Assistant Examiner(s): KENNEDY, ADRIAN L
• Application Number:
  US12/003,640
• Publication Number:
  US 20090171879A1
• Time in Patent Office: 1,544 Days
• Field of Search: None
• US Class Current: 706/21
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling

Abstract

In certain example embodiments of this invention, there is provided systems and/or methods for processing BAM-related data to predict when events of interest are about to happen and/or to identify the root causes of, or at least data correlated with, such events of interest. In certain example embodiments, key performance indicators (KPIs) are gathered and gardened. The gardening process may identify KPI values of interest (e.g., based on a Z-factor analysis thereof across one or more collection intervals). The gardened KPIs may be processed using a time-series transform (e.g., a Fast Fourier Transform), matched to one of a plurality of predefined waveforms, and fed into a dynamic Naïve Bayesian Network (NBN) for prediction. The gardened data also may be used to determine the relevance of the KPI for root causes of problems (e.g., based on a chi-square analysis).

Claims

1. A method of analyzing data in a business processing management environment, the method comprising:

identifying a plurality of performance indicators, each said performance indicator being related to a process and/or system component;

defining at least one rule to monitor at least one said performance indicator of a process and/or system component;

creating a prediction template to identify at least one rule on which to create a prediction, the prediction template including a past interval indicative of an amount of data to be analyzed before making the prediction, a future interval indicative of how far into the future the prediction is to be made, and an accuracy threshold indicative of a probability level to be reached in making the prediction;

gathering data for each said performance indicator over a plurality of collection intervals;

gardening the gathered data to discard gathered data within a normal operating range for a given collection interval;

applying a time-series transform to the gardened data to normalize any variations in collection intervals;

feeding the transformed gardened data into a dynamically updatable Naï

ve Bayesian Network (NBN) such that an entry is created for the transformed gardened data when the NBN does not include an entry for the transformed gardened data, and such that an already existing entry for the transformed gardened data is updated when the NBN includes an entry corresponding to the transformed gardened data;

making the prediction and determining an accuracy thereof using probabilities computed by the NBN; and

updating a relevance value associated with each performance indicator in a rule using the gardened data for root cause analysis,

wherein the gathering and the making of the prediction are performed substantially in real-time.

2. The method of claim 1, further comprising calculating a mean and standard deviation for each performance indicator over each collection interval,

wherein the gardening of the gathered data is performed in dependence on a comparison between the performance indicator and the mean and standard deviation thereof.

3. The method of claim 2, wherein the gathered data is not discarded when it is at least two standard deviations from the mean.

4. The method of claim 1, further comprising matching the transformed gardened data to one of a plurality of predefined waveforms;

and

replacing the transformed gardened data with the matching waveform.

5. The method of claim 1, wherein the time-series transform is a Fast Fourier Transform.

6. The method of claim 1, further comprising updating the relevance value associated with each performance indicator based on a chi-square analysis of performance indicators associated with the rule and performance indicators not associated with the rule.

7. The method of claim 1, further comprising gardening based on a chi-square relevance analysis.

8. The method of claim 1, further comprising when a prediction is made and the accuracy thereof is at least equal to the accuracy threshold of the prediction template, raising an alert.

9. The method of claim 8, further comprising generating at least one of an email message, an SMS message, and a notification window on a terminal when an alert is raised.

10. The method of claim 1, further comprising adjusting system resources and/or business processing in dependence on the prediction.

16. The system of claim 1, wherein the relevance value associated with each performance indicator is updated based on a chi-square analysis of performance indicators associated with the rule and performance indicators not associated with the rule.

11. A system of analyzing data in a business processing management environment, comprising:

a performance indicator storage location for storing a plurality of performance indicators, each said performance indicator being related to a process and/or system component;

a rules storage location including at least one rule for monitoring at least one said performance indicator of a process and/or system component;

a prediction templates storage location including at least one prediction template for identifying at least one rule on which to create a prediction, each said prediction template including a past interval indicative of an amount of data to be analyzed before making the associated prediction, a future interval indicative of how far into the future the associated prediction is to be made, and an accuracy threshold indicative of a probability level to be reached in making the associated prediction;

a connection to a data stream, the data stream including data for each said performance indicator over a plurality of collection intervals;

a gardening module configured to garden the data in the data stream to discard data within a normal operating range for a given collection interval;

a time-series transformation engine configured to apply a time-series transform to the gardened data to normalize any variations in collection intervals;

a dynamically updatable Naï

ve Bayesian Network (NBN) configured to receive the transformed gardened data such that an entry is created for the transformed gardened data when the NBN does not include an entry for the transformed gardened data, and such that an already existing entry for the transformed gardened data is updated when the NBN includes an entry corresponding to the transformed gardened data; and

a prediction engine configured to make the prediction and to determine an accuracy thereof using probabilities computed by the NBN, and further configured to update a relevance value associated with each performance indicator in a rule using the gardened data for root cause analysis,

wherein the gardening module and the prediction engine operate substantially in real-time.

12. The system of claim 11, wherein the gardening module is further configured to calculate a mean and standard deviation for each performance indicator over each collection interval, and

wherein the gardening of the gathered data is performed in dependence on a comparison between the performance indicator and the mean and standard deviation thereof.

13. The system of claim 12, wherein the gathered data is not discarded when it is at least two standard deviations from the mean.

14. The system of claim 11, further comprising a plurality of predefined waveforms, the transformed gardened data being matched to one of said waveform prior to its inclusion in the NBN.

15. The system of claim 11, wherein the time-series transform is a Fast Fourier Transform.

17. The system of claim 11, wherein the gardening module is configured to garden data based on a chi-square relevance analysis.

18. The system of claim 11, further comprising an alert module activatable when a prediction is made and the accuracy thereof is at least equal to the accuracy threshold of the prediction template.

19. The system of claim 18, wherein the alert module is configured to generate at least one of an email message, an SMS message, and a notification window on a terminal.

20. The system of claim 11, further comprising programmed logic circuitry for adjusting system resources and/or business processing in dependence on the prediction.