SEARCH

SEARCH BY CITATION

Abstract

The problem of synthesizing an optimal low risk reaction network from unreliable reactions is formulated. An expected cost criterion function for decision-making is derived both for the case of a continuously operating network of reactions and for the case of reaction paths that involve batch reactions. It is shown that in general the decision space has a nonserial structure and the search for the optimal path will involve use of network search methods. An example is given where the optimal reaction path is synthesized for deoxyribonucleic acid. This example demonstrates the decision-making strategy for a class of bath reaction systems.