Abstract: It is described a method for monitoring an optical communications system comprising at least one optical channel connecting a transmitter and a receiver. The method comprises: measuring, at the receiver, a transmission parameter of the optical channel for a pre-defined measuring time interval; on the basis of the measurements of the transmission parameter in the time interval, checking the presence of at least one anomaly in the measurements, the at least one anomaly being indicative of a subsequent failure of the system; and in the presence of the at least one anomaly, applying an identification algorithm to the measurements, the algorithm comprising a classifier, wherein the classifier is configured to, on the basis of the measurements, identify a cause of the failure, the classifier being based on a machine learning technique.