Publications / 2019 Proceedings of the 36th ISARC, Banff, Canada

Improving the Quality of Event Logs in the Construction Industry for Process Mining

Liyuan Chen, Seokyoung Kang, Shahin Karimidorabati and Carl Haas
Pages 804-811 (2019 Proceedings of the 36th ISARC, Banff, Canada, ISBN 978-952-69524-0-6, ISSN 2413-5844)
Abstract:

Process mining is an emerging tool kit to discover, analyze, and improve workflows. In the construction industry, projects can be large and unique in terms of costs and durations. In such projects, commercial software that support analyses and decision making may be helpful. Process mining software allow construction companies to quickly discover benchmark workflows, conduct conformance checking, and analyze root causes. However, engineers who initially format the event logs, and users who use the event logs for analyses may not share the same knowledge or values. Due to the lack of understanding, knowledge, or experience in these respective domains, companies may not have documented the event logs in the most practical way. With the poorly structured event logs, the software may not be able to provide the most accurate analyses. Therefore, there is a need to pre-process event logs so as to improve their quality. This paper examines a case study on Engineering Change Request (ECR) in the construction industry to explain the importance of pre-processing the event logs before importing them into the commercial process mining software. For large complex projects, the improved quality of event logs will reduce confusion between engineers and analyzers and improve the accuracy of the analysis results produced by the process mining software.

Keywords: Process mining; Event log; Pre-processing