A SDG-based reasoning procedure is presented in this paper to qualitatively predict all possible symptom patterns and also their progression sequences caused by fault propagation in any given process system. These intrinsic features of the symptom evolution behaviors are then captured with IF-THEN rules in a two-layer fuzzy inference system. The proposed diagnostic system can be used to identify not only the locations of fault origins but also their magnitude levels with relatively high resolution. Numerical simulation studies have been carried out to verify the feasibility and effectiveness of the proposed approach. Crown Copyright (C) 2008 Published by Elsevier Ltd. All rights reserved.