Anticipating Data Inaccuracy Consequences in Business Processes: an Empirical Study

In today’s data-driven business landscape, the reliability of information is paramount for making effective decisions and achieving organizational objectives. However, data inaccuracy remains a persistent challenge that can undermine the integrity of various business processes. Designing business processes and their accompanying information systems relies on the assumption that data used within the processes accurately represents reality. Unfortunately, this assumption is not always realistic, leading to potential risks that could impact both the process and the achievement of business goals. Proactively analyzing and anticipating potential data inaccuracies during design time, process designers can devise better solutions before implementation, minimizing risks at runtime. This paper evaluates the contribution of a design-time analysis of possible impacts of data inaccuracies. This analysis is built on two key concepts: 1. Data Inaccuracy Awareness (DIA), which establishes whether, at any given moment, one can be confident that data values accurately reflect the corresponding real-world values. 2. Inter-instance data impact, which captures the potential data impacts among different instances of the same process, thus scopes potential impacts of data errors. The paper reports a study that examined the effectiveness of this analysis, with the participation of experienced process designers. The study provided valuable insights into the applicability of the analysis and its potential contribution for addressing potential data-related challenges during design time.