Risk-Aware Business Process Management

Home Background Activities Outputs Participants Get Involved Acknowledgement Contact Us

Several on-going research activities that related to the risk-aware business process management project are detailed below.

Risk Visualisation

Given the massive amount of process-related data collected in today's systems, we seek to develop a framework to allow stakeholders (including business analysts and managers) to visualize their data in a manner that facilitates effective extraction of useful insights. In particular, we aim to generate animated movies based on historical data in order to improve the way in which risky events that could have happened in the past can be easily identified. Moreover, we aim to provide a customizable framework that can generate various types of animations using the same historical data to support visualization of processes from multiple perspectives simultaneously.

Run-time Risk Monitoring and Analysis

Regardless of how well a business process is designed, there are certain risks that cannot be eliminated completely due to the nature of the errors (e.g. incorrect payroll date entered), or the implementation of mediocre risk counter-measures techniques (due to budget and/or time constraints) which allow certain risks to still be manifested in the system.

Our research in runtime risk monitoring and analysis aims at addressing risks related to run-time business process execution. The idea here is that if appropriate alerts can be issued early in the process as the system goes into some pre-determined risky situations, we may be able to prevent the ultimate (and potentially catastrophic) risk event(s) from happening.

A preliminary result of this research has been published at the CoopIS 2011 conference.

Post-execution Risk Mining

In the field of business process automation, the execution traces of process instances are logged. This log information may prove valuable for verifying various properties of the executed process instances, e.g. for obtaining some statistical information about the frequency of process instances entering some pre-determined risky states and how often the risk events eventually manifested themselves, or for discovering hidden risks in process execution log which were not anticipated during design-time.

Mining Process Risks and Resource Profiles

Risk management and business process management are major concerns of corporate and government organisations. Business processes are exposed to different risks. Human resources are often involved in business processes and their actions can be a source of process risks. Information about processes is often recorded in event logs. We propose methods for identifying process risks and for mining resource profiles from event logs.
More details about this research can be found here