Assessing and improving measurability of process performance indicators based on quality of logs. Issue 103 (January 2022)

• Record Type: Journal Article
• Authors: Cappiello, Cinzia; Comuzzi, Marco; Plebani, Pierluigi; Fim, Matheus
• Is Part Of: Information systems. Issue 103(2022)
• Publication Date: 2022-01
• Subjects: Business process -- Event log -- Data quality assessment -- Data quality improvement; Database management -- Periodicals; Electronic data processing -- Periodicals; Bases de données -- Gestion -- Périodiques; Informatique -- Périodiques; Database management; Electronic data processing; Periodicals; 005.7
• DOI: 10.1016/j.is.2021.101874
• Languages: English
• ISSNs: 0306-4379

Abstract

Abstract: The efficiency and effectiveness of business processes are usually evaluated by Process Performance Indicators (PPIs), which are computed using process event logs. PPIs can be insightful only when they are measurable, i.e., reliable. This paper proposes to define PPI measurability on the basis of the quality of the data in the process logs. Then, based on this definition, a framework for PPI measurability assessment and improvement is presented. For the assessment, we propose novel definitions of PPI accuracy, completeness, consistency, timeliness and volume that contextualise the traditional definitions in the data quality literature to the case of process logs. For the improvement, we define a set of guidelines for improving the measurability of a PPI. These guidelines may concern improving existing event logs, for instance through data imputation, implementation or enhancement of the process monitoring systems, or updating the PPI definitions. A case study in a large-sized institution is discussed to show the feasibility and the practical value of the proposed framework. Highlights: Define PPI measurability in respect of the data quality of process event logs. Propose a model for assessment of event log data quality in respect of PPI definitions. Propose guidelines to improve the process monitoring infrastructure to achieve higher PPI measurability. A case study in the service industry demonstrates the feasibility and value of the proposed framework.