Using Contextualized Activity-Level Duration to Discover Irregular Process Instances in Business Operations(2/2)

Effective time management has been one of the most crucial characteristics of a successful business. For most businesses, it has been an area that could always be improved. Irregularities in execution duration of business processes impede corporate agility and could cause severe consequences, such as project failure and financial losses. Efficient managers must to constantly identify potential irregularities in duration to foresee and avoid process glitches. In the first year, this study proposes a k-Nearest Neighbor method for systematically detecting irregular process instances in a business by using a comprehensive set of activity-level durations. These activity-level durations are execution, transmission, queuing, and procrastination durations. Moreover, agents, customers, and other variables influence the progresses of processes. In the second year, contextual information is presented using fuzzy values. The values and corresponding membership functions are used to adjust the durations of each activity. This method will be applied to the system logs of a medium-sized logistics company to identify irregularities. Senior managers in the company will review the correctness of the identified process irregularity to evaluate the accuracy of the proposed method. The result will also be used in the company to improve process management.