Survey and cross-benchmark comparison of remaining time prediction methods in business process monitoring

Predictive business process monitoring methods exploit historical process execution logs to generate predictions about running instances of a process, including predictions of the remaining cycle time of running cases of a process. A number of approaches to tackle this latter prediction problem have been proposed in the literature. However, due to differences in the experimental setups, choice of datasets, evaluation measures and baselines, the relative performance of various methods remains unclear. This article presents a systematic review and taxonomy of methods for remaining time prediction in the context of business processes, as well as a cross-benchmark comparison of 16 methods based on 16 real-life datasets