Extensive amount of detection and analysis of healthy working circumstance of Induction Motors (IM) are proposed and implemented in literature. This paper considers, heuristic approach assisted adaptive filter is recommended to detect and analyze the vibration signal of IM. In this work, a digital Infinite Impulse Response (IIR) filter is designed using recent heuristic algorithm known as the Firefly Algorithm (FA). During this study, a model squirrel-cage induction motor running at various load conditions, such as 25%, 50%, 75% and 100% are considered and the variation of the vibration level with respect to the variation in load is analyzed using the proposed procedure. The performance of the FA based IIR filter (FA-IIRF) is then validated against the IIR filter designed using PSO and BA based approaches. This experimental study confirms the capability of FA-IIRF in detecting the common induction motor faults like, vibration due to weak bearing and vibration due to the stator winding. Finally, the statistical significance of the FA-IIRF is confirmed using the Wilcoxon's rank-sum test. This test also proves that, FA based approach is better than PSO and BA based approach. © 2017 IEEE.