How planning techniques can help process mining: The conformance-checking case

Modern organizations execute processes to deliver product and services, whose enactment needs to adhere to laws, regulations and standards. Conformance checking is the problem of pinpointing where deviations are observed in the process event data. Literature proposes solutions for the conformance-checking problem that, in fact, are implementations of planning algorithms built ad-hoc for the specific problem. Unfortunately, in the era of big data, these ad-hoc implementations do not scale sufficiently compared with robust, well-established planning systems. Furthermore, their ad-hoc nature does not allow for seamlessly plugging in new outperforming planning algorithms or heuristics, causing a massive amount of work to be necessary to incorporate and evaluate alternatives. This paper summarizes our last results on how instances of the conformance checking problem can be represented as classical planning problems in PDDL for which planners can find a correct solution in a finite amount of time. If conformance checking problems are converted into planning problems, one can seamlessly update to the recent versions of the best performing automated planners, with evident advantages in term of versatility and customization. Experiments with three different planners highlight this versatility. Furthermore, by employing several processes and event logs of different sizes, we show how our planning-based approach outperforms existing approaches of several order of magnitude and, in certain cases, carries out the task while existing approaches run out of memory.