Mirror, mirror on the wall, can I count on you at all? Exploring data inaccuracy in business processes

Abstract

Information systems in general and process aware information systems in particular support the execution of business processes. This support is based on the assumption that the information system truly reflects the state of the domain where the process takes place. Based on this assumption, humans do not need to directly "sense" the state of affairs. Rather, decisions are made based on the state as reflected in the information system. This paper explores the situation where this assumption does not hold, namely, the situation of data inaccuracy. In particular, it formalizes data inaccuracy and addresses three questions: (a) how is data inaccuracy manifested in a process? (b) what are the expected results of data inaccuracy? (c) how can robustness to data inaccuracy be increased? The understanding gained with respect to these questions should form a basis for designing processes to be more robust, avoiding problems due to data inaccuracy. © 2010 Springer-Verlag Berlin Heidelberg.