Conformance checking

[1] It is used to check if the actual execution of a business process, as recorded in the event log, conforms to the model and vice versa.

For instance, there may be a process model indicating that purchase orders of more than one million euros require two checks.

Another example is the checking of the so-called “four-eyes” principle stating that particular activities should not be executed by one and the same person.

By scanning the event log using a model specifying these requirements, one can discover potential cases of fraud.

[2] Conformance checking techniques take as input a process model and event log and return a set of differences between the behavior captured in the process model and the behavior captured in the event log.

Some techniques may also produce a normalized measures (between 0 and 1) indicating to what extent the process model and the event log match each other.

The interpretation of non-conformance depends on the purpose of the model: The purpose of conformance checking is to identify two types of discrepancies: There are broadly three families of techniques for detecting unfitting log behavior: replay, trace alignment and behavioral alignment.

Hence, these methods might not identify the minimum number of errors that can explain the unfitting log behavior.

If for example four tasks can occur only in a fixed order in the process model (e.g. [A, B, C, D]), but they can occur concurrently in the log (i.e. in any order), this difference cannot directly detected by trace alignment, because it cannot be observed at the level of individual traces.

If the process model can replay the negative events, it means that there is behavior captured in the process model that is not captured in the log (since the negative events correspond to behavior that is never observed in the log).

For a process model, such a matrix can also be derived on top of the execution sequences by using the play-out technique.

Although the token-replay technique is efficient and easy to understand, the approach is designed for Petri net notation and doesn't consider the suitable path generated by the model for the unfit cases.

Alignments were introduced to solve the limitations and is considered a highly accurate conformance checking technique and can be applied for any process modeling notation.

[8] The idea is that the algorithm performs an exhaustive search to find out the optimal alignment between the observed trace and the process model.