However, ABPD accelerates process discovery and identifies previously unseen process patterns. This is why ABPD is best combined with techniques from social BPM and BPA for the masses. Of course, ABPD does not capture informal human tasks that are not automated and represented as events in an audit trail or transaction log. It’s all about capturing what has happened in a real-life process. Process modeling and workflow applications have become more an more important during last decade. Moreover, ABPD delivers information on bottlenecks, variances, root causes and the distribution of the process instances, thus enabling meaningful simulation. How the above mentioned principles can be integrated together and how the standard approach in process specification might be extended with the spatial dimension to make business process models more natural and understandable are discussed. The underlying techniques are strong enough so that users don’t have to specify a process model it is created from scratch by pattern recognition. ABPD techniques start from event logs (audit trails, messages, transactions, databases and so forth), and try to discover patterns to fit a process model to the information obtained from the events. Because of the quantitative nature of this data, the derived process models enable rich and interactive analysis. By observing the details of how processes are executed in the supporting technology, ABPD uses unstructured, event-level data to automatically build process definitions and models, and explore process variations.
Modeling can be done by hand, so there may be no cash-flow impact, unlike ABPD, where a tool is necessary. One major benefit of ABPD is process discovery across the “white space,” the process knowledge gap that exists between departments and functions and at the edges of processes. Automated business process discovery (ABPD) as a complementary approach overcomes many of these shortcomings to create a business process model at a fraction of the time and cost involved in the traditional way.