Gears are machine elements that transmit motion by means of successively engaging teeth. In purely scientific terms, gears are used to transmit motion. A faulty gear is a matter of serious concern as it affects the functionality of a machine to a great extent. Thus it is essential to diagnose the faults at an initial stage so as to reduce the losses that might be incurred. This necessitates the need for continuous monitoring of the gears. The vibrations produced by gears from good and simulated faulty conditions can be effectively used to detect the faults in these gears. The introduction of Variational Mode Decomposition (VMD) as a new signal pre-processing technique along with the different decision trees have provided good classification performance. VMD allows decomposition of the signal into various modes by identifying a compact frequency support around its central frequency, such that adding all the modes reconstructs the original signal. Alternating direction multiplier method (ADMM) is used by VMD to find the intrinsic mode functions on central frequencies. Meaningful statistical features can be extracted from VMD processed signals. J48 decision tree algorithm was used to identify the useful features and the selected features were used for classification using the decision trees namely, Random Forest, REP Tree and Logistic Model Tree algorithms. The performance analyses of various algorithms are discussed in detail. Copyright © 2014 Tech Science Press.