An Approach to Data-Driven Process Discovery in the Cost Estimation Process of a Construction Company

Abstract

This Work Examines The Potential Of Process Mining In The Architecture, Engineering, Construction, And Operation (AECO) Industry, Where Process Mining Is Rarely Applied. The Main Reason Is That Standardised Processes Are Hardly Ever Performed Due To The Complexity Of Projects. To Address This Application Gap, The Software-Supported Cost Estimation Process For A Tender In A German Construction Company Is Examined. For This Purpose, Data Sets From Three Different Projects Are Exported From The Software RIB ITWO And Analysed With Regard To Process Discovery. Investigations Are Carried Out From The Control Flow, Case, And Organisational Perspectives. A Particular Problem In The Analysing Part Is The Currently Inadequate Quality Of The Data Sets To Be Examined. Due To Data Quality Issues, The Consideration Of Temporal Aspects Is Rather Not Possible. Therefore, Consistent And Appropriate Forms Of Logging Must Be Implemented In Software Systems That Are Used To Support the AECO industry. For The Present Use Case, An Automated Working Script Was Created To Process The Log Data To Fix Further Quality Issues And Meet Process Mining Requirements. The Results Show That It Is Essential To Establish Standardised Language Rules That Are Generally Valid Throughout The Industry And That Are Subsequently Referenced By All Utilized Software Systems. This Enables Comparative Analyses Across Projects And Companies To Make Process Mining Methods Become Routinely And Profitably Applicable In The AECO Industry.