System and method for predictive process management

• US 7,680,752 B1
• Filed: 01/06/2006
• Issued: 03/16/2010
• Est. Priority Date: 01/06/2005
• Status: Active Grant

• Current Assignee: Parasoft Corporation
• Original Assignee: Parasoft Corporation
• Inventors: Dominguez, Daniel John, Clune, James Edmond III
• Primary Examiner(s): Holmes; Michael B
• Application Number:
  US11/326,930
• Time in Patent Office: 1,530 Days
• Field of Search: 706/45
• US Class Current: 706/45
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling

Abstract

An integration platform for managing long term processes. The platform includes a runtime engine that manages the process instances and communication with the processes to provide Web services. The runtime engine includes an execution engine module and a prediction module. The prediction module generates predictions for executing processes for inbound and outbound messages and eventual process outcomes. The predictions may be utilized when requisite data is unavailable or under defined circumstances. The prediction module utilizes process history, process description and process state data to generate relevant prediction data. The prediction module may include a learning module that applies analytical algorithms to the process history and descriptions to create models for use by the prediction module.

Claims

1. A process management system comprising:

an execution engine to execute a process, the execution engine being configured to provide a process history including an audit trail, a process state and a process description;

a prediction module in communication with the execution engine, the prediction module being configured to provide a process prediction to the execution engine based on the process history, the process state and the process description; and

a learning module in communication with the prediction module to process one or more of the process history and the process description and generate a learning model for the prediction module, wherein the learning model includes a decision tree for predicting message events or a process state, where nodes of the decision tree involve preceding message events or values within the process state.

2. The process management system of claim 1, wherein the process prediction includes one or more of a message event, a process outcome and a process definition.

3. The process management system of claim 1, wherein the decision tree is logic-based including use of one or more of first order logic and predicate calculus, and includes probabilistic elements including Bayesian probabilities.

4. The process management system of claim 1, further comprising:

a domain-specific knowledge base accessible to the prediction module.

5. The process management system of claim 1, further comprising:

a process manager in communication with the execution engine to retrieve a process description.

6. The process management system of claim 1, further comprising:

a process instance database for storing processes.

7. The process management system of claim 1, further comprising:

a graphical user interface to display a process via multiple views.

8. The process management system of claim 7, wherein each view shows a representation of a same data model.

9. The process management system of claim 7, wherein the multiple views are a model view, an outline view, a source view and/or an attribute view.

30. The process management system of claim 1, wherein the process prediction includes a reliability rating or confidence level to be used as conditional information for determining utilization of the process prediction.

31. The process management system of claim 1, wherein the process prediction is related to message events and includes predictions of type or predictions of content of message that will be received.

32. The process management system of claim 1, wherein the prediction module is configured to predict a next message event or a next message content by considering what happened in the process histories of similar processes at a similar point in the execution of a process.

33. The process management system of claim 1, wherein the audit trail is a XML file and each event is represented as an element in the XML file.

10. A method for managing processes comprising:

executing a process by an execution engine;

generating a process history including an audit trail, and a process state;

processing one or more of the process history, process state and the process description, and generate a learning model for the prediction module, wherein the learning model includes a decision tree for predicting message events or a process state, where nodes of the decision tree involve preceding message events or values within the process state; and

processing the process history, the process state and the process description by the prediction module to provide the execution engine with a process prediction.

11. The method for managing processes of claim 10, wherein the process prediction includes a reliability rating or confidence level to be used as conditional information for determining utilization of the process prediction.

12. The method of managing processes of claim 10, wherein the process prediction includes one or more of a message event, a process outcome and a process definition.

13. The method of managing processes of claim 10, further comprising:

applying one or more of analytic and inductive learning to generate the process prediction.

14. The method of managing processes of claim 10, further comprising:

applying a domain-specific knowledge base to generate the process prediction.

15. The method of managing processes of claim 10, further comprising:

applying a process prediction if a time period has elapsed.

16. The method of managing processes of claim 10, further comprising:

tracking a virtual state.

17. The method of managing processes of claim 10, further comprising:

generating virtual messages.

18. The method of managing processes of claim 10, wherein a granularity of the process prediction ranges from a technical level to a business level.

19. The method of managing processes of claim 10, wherein the process is a business process execution language (BPEL) process.

20. The method of managing processes of claim 10, wherein the process prediction ranges from a micro-scale to a macro-scale.

21. The method of managing processes of claim 10, wherein the execution engine maintains a virtual state.

22. The method of claim 10, further comprising:

retrieving the process from a stored state.

23. The method of claim 10, wherein the process is part of a long term transaction.

24. The method of claim 10, further comprising:

displaying multiple views of a data model representing the process through a graphical user interface.

25. The method of claim 24, wherein the multiple views are a model view, outline view, a source view and an attribute view.

26. The method of claim 24, further comprising:

updating each of the multiple views automatically upon receiving an input.

27. A method comprising:

displaying a source of a process including an audit trail through a graphical user interface;

displaying a model view of the process through the graphical user interface simultaneous with the source;

displaying an attribute view of the process through the graphical user interface simultaneous with the model view; and

updating a data model based on input received in any view, wherein the data model includes a decision tree for predicting message events or a process state.

28. The method of claim 27, further comprising:

displaying an outline view of the process through the graphical user interface simultaneous with the attribute view.

29. The method of claim 27, further comprising:

updating each view automatically based on the input.