Publications / 2024 Proceedings of the 41st ISARC, Lille, France
Despite current digital transformation attempts, productivity in the construction industry has remained stagnant for decades. Construction organizations generate vast amounts of data from their day-to-day operations. However, these valuable digital footprints frequently remain underutilized and isolated in different IS (Information Systems). To tackle this issue, process mining, a novel and robust technology, provides organizations with the means and methods to automatically monitor and improve the efficiency of business processes by exploiting event process data. Since construction projects are process-heavy, process mining adoption becomes essential to overcome productivity stagnation. To enable process mining capabilities, process data is required to be captured in the form of event logs. In this regard, the existing body of knowledge in the construction domain yet lacks a system-agnostic framework for event log generation that considers current standard data schemas such as XES (eXtensible Event Stream) and OCEL (Object Centric Event Logs) to ensure event logs soundness and machine readability. Therefore, to address this limitation, this work aims to facilitate the production of event logs with proper syntax and semantics by: (i) developing an ETL (Extract, Transform, Load) framework to harness process data from IS; (ii) analyzing and describing the XES and OCELs relational data structures; (iii) providing a comparative analysis of both event log data standards. The results include a functional demonstration for constructing these data schemas applied to the Change Order Management Process (COMP) implemented in a commercial project. Construction change orders have the potential to either contribute to construction projects success or yield to failure.