Relevant and reliable data from the surroundings in addition to the vibration measurements are required to achieve effective diagnosis and prognosis when a vibration-based maintenance (VBM) policy is used.

The main conclusion is that the variation in the machine load and speeds influences the amplitudes of rolling element bearing defect frequencies. Therefore, changes in the machine speed and load should be considered when interpreting vibration spectra. This will improve the effectiveness and accuracy of fault diagnosis and prediction of the time to maintenance action.

The paper classifies stoppage times and highlights the reasons behind them. Also, it discusses the consequential economic losses incurred by unplanned stoppages. During the period covered by this study (fifty-eight days), the total stoppage time was about 76.6 days and caused appreciable economic losses of about 2,3 millions SEK.