Expert systems with on-line and real-time capabilities are required wh
en the application is to be to a technical process whose states change
with time; so also the contents of the expert system s knowledge-base
. Also, rule-based systems are not powerful enough to satisfy the requ
irements for an application to be useful. New paradigms are necessary
to cope with the problems. Integration of qualitative and quantitative
modelling with respect to expert system applications to large-technic
al processes is a must. As processes are event-based, so must be the e
xpert system applied to it. Moreover, the paper shows that event orien
tation has a natural software paradigm as a counterpart: object orient
ation. A method as to how the dynamical changes in the process are ref
lected in a dynamical (time-dependent) knowledge-base is also shown.