A combined passive/active fault detection and isolation approach is presented based on the use of data-driven models to detect faults, and Active Functional Testing (AFT) to isolate them in air handling units (AHU). A model developed in Modelica allows for simulation of the dynamic behavior of the system under different operating conditions. Four case studies are presented for the supply air temperature sensor under different sensor biases. Two steps are followed in the proposed strategy to identify the cause of the abnormal operation in the system: the first step is to detect the fault or a set of potential faults based on the principal components analysis (PCA) method; the second step is to isolate the effective fault among possible candidates by applying AFT. Simulation results and site experiments show that this approach is able to effectively detect and isolate faults in AHUs. Crown Copyright (C) 2015 Published by Elsevier B.V. All rights reserved.