Local Concurrency Detection in Business Process Event Logs

Abstract

Process mining techniques aim at analysing records generated during the execution of a business process in order to provide insights on the actual performance of the process. Detecting concurrency relations between events is a fundamental primitive underpinning a range of process mining techniques. Existing approaches to this problem identify concurrency relations at the level of event types under a global interpretation. If two event types are declared to be concurrent, every occurrence of one event type is deemed to be concurrent to one occurrence of the other. In practice, this interpretation is too coarse-grained and leads to overgeneralization. This paper proposes a finer-grained approach, whereby two event types may be deemed to be in a concurrency relation relative to one state of the process, but not relative to other states. In other words, the detected concurrency relation holds locally, relative to a set of states. Experimental results both with artificial and real-life logs show that the proposed local concurrency detection approach improves the accuracy of existing concurrency detection techniques.

Generalization in Automated Process Discovery: A Framework based on Event Log Patterns

“…A local concurrency oracle [26] addresses this shortcoming by finding concurrency relations for pairs of activities at a given execution context. Then, requires three elements, two activity labels x, y and an execution context C, i.e., (x, y, C) = 1 if x and y are concurrent at C, and (x, y, C) = 0 otherwise.…”

“…Then, requires three elements, two activity labels x, y and an execution context C, i.e., (x, y, C) = 1 if x and y are concurrent at C, and (x, y, C) = 0 otherwise. For the sake of brevity, the details of the local oracle presented in [26] are not shown in this paper, and the oracle is used as a blackbox.…”

“…Hence, a local concurrency oracle can identify less over-generalizing concurrency relations. However, the technique presented in [26] has been shown to be less scalable than global oracles and cannot handle large datasets.…”

“…In Section 5, we use the alpha+ oracle [23] with a frequency filter as a representative global concurrency oracle and [26] as a local concurrency oracle.…”

Semi-supervised Log Pattern Detection and Exploration Using Event Concurrence and Contextual Information

“…3 Partial orders of events may be obtained from totally ordered traces. A few works [15,17,18] assume to have an oracle that indicates the set of activities that are concurrent or unordered and use this oracle to convert totally ordered events into partial orders. Such an oracle could be obtained by interviewing domain experts or be computed from event logs [14].…”