Page 352 -

334 13 Cartography and Navigation

Fig. 13.10 The fuzzy model discovered earlier (cf. Fig. 13.7(a)) is used to replay the event log.
The animation reveals the problem that many reviewers do not provide a report in time. As a result,
the editor of the journal cannot make a final decision and needs to invite additional reviewers. There
is long queue of work items waiting for a decision and many pending invitations

the prediction capabilities of a navigation device. Moreover, using different annota-
tions, other kinds of predictions can be made. For instance, the transition system can
be annotated with cost information to predict the total or remaining costs. Similarly,
the outcome of a process or occurrence of an activity can be predicted.
Alternative approaches based on regression analysis, short-term simulation, or
decision tree learning can be used to predict properties such as the remaining flow
time of a running case. This illustrates that process mining can be used to extend
information systems with predictive analytics.

13.2.3 Guidance Rather than Control

Car navigation systems provide directions and guidance without controlling the
driver. The driver is still in control, but, given a goal (e.g., to get from A to B as
fast as possible), the navigation system recommends the next action to be taken. In
Sect. 9.5, we showed that predictions can be turned into recommendations. Recom-
mendations are given with respect to a goal, e.g., to minimize costs, to minimize
the remaining flow time, or to maximize the likelihood of success. Such a goal is
operationalized by defining a performance indicator that needs to be minimized or