Measuring process model performance and enforcing process performance policy

Patent metadata

Field	Value
Publication number	US-9600795-B2
Application number	US-201213442462-A
Country	US
Kind code	B2
Filing date	Apr 9, 2012
Priority date	Apr 9, 2012
Publication date	Mar 21, 2017
Grant date	Mar 21, 2017

Abstract

Common sub-process patterns in a plurality of deployed process models may be discovered, and performance measures associated with the sub-process patterns may be computed based on runtime events of the deployed process models. Positive or negative performance patterns among sub-process patterns may be identified and used for creating new process models or improving existing process models.

First claim

We claim:

1. A method for process model performance measurement and process performance policy enforcement, comprising:
receiving a deployed process model;
discovering a common sub-process pattern in the deployed process model that is in common with at least one previously discovered sub-process pattern in a previously deployed process model;
adding the deployed process model to a member list of the common sub-process pattern, if the deployed process model is not already in the member list, the member list of the common sub-process pattern comprising at least a plurality of process patterns that includes the common sub-process pattern, the common sub-process pattern associated with activities in the process patterns that reify the common sub-process pattern, the member list further comprising information that tells where a given pattern appears;
receiving a signal of one or more events associated with a runtime instance of the deployed process model, the signal of one or more events describing one or more of start, stop, failure, exception, and configured events occurring in an execution of said runtime instance of the deployed process model, the signal of one or more events stored in an event data store;
responsive to determining that the deployed process model or a sub-process within the deployed process model that caused the one or more events to be generated is in the member list of a common sub-process pattern, associating one or more performance dimensions impacted by the one or more events with the common sub-process pattern;
computing performance measurement associated with the discovered common sub-process pattern based on the received one or more events;
based on the performance measurement, identifying a composition of two or more services that individually produce positive performance but when composed together produce negative performance,
wherein generating of the one or more events are configurable to be turned on and off at least for individual process models or instances of process models conforming to a given pattern; and
displaying via a dashboard user interface, the common sub-process pattern and the performance measurement,
wherein a plurality of common sub-process patterns are discovered and saved with respective performance measurements for creating and updating of new process models, performance of the new process models further predicted based on the respective performance measurements of the plurality of common sub-process patterns,
wherein a sub-process pattern data store is rebuilt at set intervals.

2. The method of claim 1 , further including discovering existence of negative performance pattern or positive performance pattern or combinations thereof in a newly deployed model based on the computed measurement.

3. The method of claim 1 , wherein the method is performed for a plurality of deployed process models, wherein a plurality of sub-process patterns are discovered and the performance measurement is computed for each of the plurality of discovered sub-process patterns.

4. The method of claim 3 , wherein a list of the plurality of sub-process patterns and associated performance measurement is presented for creating a new process model or improving an existing process model or combinations thereof.

5. The method of claim 3 , further including predicting performance of a new process model based on the plurality of sub-process patterns and associated performance measurement.

6. The method of claim 3 , further including enforcing one or more defined policies on the deployed process model based on the plurality of sub-process patterns and associated performance measurement.

7. The method of claim 3 , wherein the events are configurable.

8. The method of claim 1 , wherein the performance measurement includes a performance pattern describing behavior being exhibited by the sub-process patterns with respect to a performance dimension.

9. The method of claim 1 , wherein the performance measurement includes probability distribution of an event recognized to impact a performance dimension of the deployed process model.

10. A system for process model performance measurement and process performance policy enforcement, comprising:
a processor;
a process analyzer operable to execute on the processor and further operable to discover a common sub-process pattern in a deployed process model that is in common with at least one previously discovered sub-process pattern in a previously deployed process model,
the process analyzer further operable to add the deployed process model to a member list of the common sub-process pattern, if the deployed process model is not already in the member list, the member list of the common sub-process pattern comprising at least a plurality of process patterns that includes the common sub-process pattern, the common sub-process pattern associated with activities in the process patterns that reify the common sub-process pattern, the member list further comprising information that tells where a given pattern appears,
the process analyzer module further operable to receive a signal of one or more events associated with a runtime instance of the deployed process model, the signal of one or more events describing one or more of start, stop, failure, exception, and configured events occurring in an execution of said runtime instance of the deployed process model, the signal of one or more events stored in an event data store,
responsive to determining that the deployed process model or a sub-process within the deployed process model that caused the one or more events to be generated is in the member list of the common sub-process pattern, the process analyzer is operable to associate one or more performance dimensions impacted by the one or more events with the common sub-process pattern, and compute performance measurement associated with the discovered common sub-process pattern based on the received one or more events,
wherein the process analyzer discovers a plurality of common sub-process patterns in a plurality of deployed process models and computes performance measurement associated with each of the plurality of common sub-process patterns, and
wherein the process analyzer identifies negative performance patterns and positive performance patterns among the plurality of common sub-process patterns based on the computed performance measurement,
wherein generating of the one or more events are configurable to be turned on and off at least for individual process models or instances of process models conforming to a given pattern
a dashboard user interface displaying the common sub-process pattern and the performance measurement,
wherein a plurality of common sub-process patterns are discovered and saved with respective performance measurements for creating and updating of new process models, performance of the new process models further predicted based on the respective performance measurements of the plurality of common sub-process patterns,
wherein a sub-process pattern data store is rebuilt at set intervals.

11. The system of claim 10 , further including storage device for storing the plurality of deployed process models, the one or more events, and the computed performance measurement associated with each of the plurality of common sub-process patterns.

12. The system of claim 10 , further including a user interface module operable to present performance alerts based on the computed performance measurement.

13. The system of claim 10 , further including a process improvement module operable to allow a user to create a new process model or improve an existing process model or combinations thereof,

Assignees

• Demuth Steve
• Heching Aliza R
• Sun Jimeng
• Diament Judah M
• IBM

Inventors

• Demuth Steve
• Heching Aliza R
• Sun Jimeng
• Diament Judah M

Classifications

• G06N5/00

  Computing arrangements using knowledge-based models · CPC title

• G06N5/04

  Inference or reasoning models · CPC title

• G06F16/951

  Indexing; Web crawling techniques · CPC title

• G06Q10/06

  Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling · CPC title

• G06Q10/067 Primary

  Enterprise or organisation modelling · CPC title