Abstract

PEREZ-ALFONSO, Damián; YZQUIERDO-HERRERA, Raykenler; PUPO-HERNANDEZ, Eudel  and  ORELLANA-GARCIA, Arturo. Process mining for process comprehension on diagnosis stage. Ing. Ind. [online]. 2018, vol.39, n.2, pp. 146-159. ISSN 1815-5936.

Process mining aims to discovering, monitoring and improving process, starting from events logged during process execution on information systems. Nevertheless, available process mining techniques for process diagnosis offer insufficient information about event log completeness or infrequent process behavior, affecting process comprehension. A new technique, named Variants Miner is presented, starting from limitations of process mining techniques regarding process comprehension. This technique aims to construct different variants of process models through process decomposition on sub-processes. An application of Variants Miner and other techniques in a real environment is presented. The obtained results show that the proposal decreases the cognitive load of the model and improve the fitness, generalization and precision, considering missing information and noise, improving process comprehension on process diagnosis stage.

Keywords : process diagnosis; process mining; process comprehension; models variants.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )