Purpose <p>In recent years, there has been a widespread use of micro-electromechanical system (MEMS) technology by researchers in the field of condition monitoring. This research article is focused on developing a MEMS framework for the condition monitoring of rolling bearings using vibration analysis. This article also addresses the technical challenges experienced during the development of the MEMS system in terms of sampling frequency and storing the acquired vibration signals at the same sampling frequency.</p> Methods <p>Vibration data is acquired using various developed MEMS configurations and standard data acquisition systems for healthy and faulty setup. The time and frequency domain signal analysis approaches are used to extract the statistical parameters, such as Kurtosis, Crest Factor, Skewness, etc., and to recognize the failure patterns from the acquired data.</p> Results <p>The comparison study results reveal that the NucleoF401RET6 + ADXL1002z configuration of MEMS setup successfully detects abnormalities in the statistical parameters. Also, in the envelope spectrums of vibration signals for NucleoF401RET6 + ADXL1002z configuration, sidebands are visible across the characteristic frequency, and their harmonics strongly indicate the presence of a fault in the inner raceway of the bearing. Further, the vibration data is acquired for bearing ball and outer raceway faults with ramp-up and ramp-down speed conditions using selected NucleoF401RET6 + ADXL1002z configurations of MEMS. Ramp-up and ramp-down analysis revealed significant fluctuations and higher amplitudes in bearing ball and outer raceway fault configurations compared to the healthy configuration.</p> Conclusion <p>The NucleoF401RET6 + ADXL1002z configurations of MEMS effectively identify a fault in the rolling bearing and are selected for diagnosis of bearing fault.</p> Graphical abstract <p></p>