AIRL LogoAIRL Research Projects


Designing Workflow Processes in Dynamic Organizational Settings

  • Investigator: Amit Deokar
  • Key Ideas:
    • Motivation: Current state-of-the-art workflow management systems focus on procedural and static process models, which are unsuitable in adaptive settings where business rules change frequently.
    • Propose using declarative AI planning and learning techniques to facilitate dynamic process model generation, on a per instance basis.


  • Deokar, A. V. “Flexibility in business process management systems: Review and open problems” (in preparation for book chapter in Business Process Management: Concepts, Technologies and Applications, (Eds.) Shankararaman, V., and Zhao, J. L., M. E. Sharpe Inc.).
  • Deokar, A. V., and El-Gayar, O. F. “Business process management systems for supporting individual and group decision making in organizational processes,” in Encyclopedia of Decision Making and Decision Support Technologies, (Eds.) Adam, F., and Humphreys, P., Idea Group Inc. (forthcoming, 2007).
  • Deokar, A. V., Madhusudan, T., Briggs, R. O., and Nunamaker, J.F., Jr. “A structured approach to designing interleaved workflow and groupware tasks,” In Proceedings of Americas Conference of Information Systems (AMCIS 2004), New York, 2004.

Process Retrieval and Reuse in Scientific Workflows

  • Investigators: Surendra Sarnikar and Amit Deokar
  • Key Ideas:
    • Apply information retrieval techniques for process retrieval.
    • Facilitate better reuse of process snippets in scientific workflows.

Workflow Instance Prediction for Resource Reservation

  • Investigators: Amit Deokar and Surendra Sarnikar
  • Key Ideas:
    • Use the information embedded in process instance logs to determine decision rules in a workflow process.
    • Use the decision rules to predict the behavior of ongoing workflow instances
    • Enable resource reservation with this prediction information to ensure better workflow quality of service.

Last Updated: 7/31/12