Page 334 -

316 12 Analyzing “Spaghetti Processes”

Fig. 12.13 Social network showing handovers between different organizational units of the AMC
hospital

ventional workflow technology is not applicable and a better understanding of the
processes is needed.
Figure 12.12 shows an example of a process model constructed for the AMC hos-
pital. The model was discovered based on event data of a group of 627 gynecological
oncology patients treated in 2005 and 2006. All diagnostic and treatment activities
have been recorded for these patients. Clearly, this is a Spaghetti process. However,
as shownin[64] it is possible to create simple models for homogeneous groups
of patients using the hierarchical clustering technique illustrated by Fig. 12.7.The
same event log also contained information about resources. For instance, Fig. 12.13
shows a social network based on this log. As in earlier examples, the social network
is based on handovers of work. However, now we do not look at individuals but at
the level of organizational units. Figure 12.13 can be used to analyze the flow of
work between different departments of the AMC hospital. For example, the social
network reveals that most handovers take place between the gynecology department
and the general clinical lab.
Experiences with process mining in several hospitals revealed important chal-
lenges when applying this new technology. The databases of hospitals contain lots
of event data. Since any event can be linked to a patient, correlation is easy. However,
for many events only the date (“31-12-2010”) is known and not the exact timestamp
(“31-12-2010:11.52”). Therefore, it may be impossible to deduce the order in which
events took place. Another problem is related to the trade-off illustrated by the den-