In order to design more operable or even optimal chemical processes, reliab
ility theory is proposed in this paper to support planning of operating pro
cedures. A three-tier reasoning framework is constructed to describe the tw
o-step quantitative planning strategy. The first step is determination of o
ptimal operating path, and the second step is generation of optimal primiti
ve operations. Based on the deficiency of the NLP method in obtaining an op
timal operating path, this paper presents the knowledge-model (K-M) method
in which expert knowledge base is combined with process dynamic model to so
lve this problem. Finally, the methodology is illustrated by an example of
raising temperature procedures for a subsystem of charge gas dehydrator reg
eneration in an ethylene plant. (C) 2000 Elsevier Science Ltd. All rights r
eserved.