This paper presents the general methodology to use spectroscopic measuremen
ts directly for on-line process monitoring and detection and diagnosis of d
isturbances. An application of the on-line monitoring of a chemical batch r
eaction using UV-visible spectroscopy is discussed in detail. Successful hi
storical batches are used to build a statistical model of the batch reactio
n. The model uses external information such as the pure spectra of the comp
ounds and their concentration profiles to improve the interpretability. Con
trol charts are developed for on-line monitoring of nea batches. It is show
n that this model is capable of detecting erroneous batches. In combination
with contribution plots,the actual cause of the disturbances can be diagno
sed.