Currently, the volume fraction of a glass fibre/matrix based composite material is being assessed only by destructive techniques. Instead of changing or destroying the structure, a new non-destructive approach based on vibration technique is proposed in this research. Further, the main objective of this paper is on the determination of fibre/matrix volume fraction using vibration analysis. A complete experimental protocol has been developed to record the vibration signal produced from experimental plates with different volume fraction. The recorded vibration signals were analyzed both in time and frequency domains. Subsequently, statistical parameter features were the extracted and associated to the volume fraction levels. Artificial Neural Network (ANN) models were then developed to classify the level of volume fraction. The classification performance of the developed network models were in the range of 80-100%. From the results, it has been observed that the network model with frequency band based features has yielded a better classification performance.