As industrialisation progresses, electric motors are increasingly utilised in manufacturing sectors, and their regular operation plays a crucial role in enhancing production efficiency, safety, and ease. Consequently, there's a growing emphasis on developing technology for monitoring the condition of electric motors. This study focuses on the analysis of common issues like rotor bar failure and eccentricity in induction motors, examining their causes, creating motor models in both normal and malfunctioning conditions through computer simulations, identifying the stator current signals, and comparing their spectra to validate the stator current data. Additionally, this research offers a dependable and efficient dataset for further analysis. The complex and fluctuating nature of the current signals in induction motors necessitates the use of advanced techniques like the tunable-Q wavelet transform (TQWT) and box dimension method for feature extraction, which is more effective in signal characterisation than other approaches. The study then explores the application of Support Vector Machines (SVM) and Artificial Neural Networks (ANN) in fault diagnosis, achieving accuracies of 91.67% and 100%, respectively. The findings indicate that ANN is superior to SVM and suggest this strategy for the automatic detection of motor faults. Implementing such intelligent systems can prevent unexpected and unplanned production interruptions caused by electric motor failures.