<p>The conveyor system is a mechanically operated tool that efficiently and rapidly transfers items and loads over an area. It powers an idler pulley with a motor, which moves the pulley down the length of the belt. This strategy decreases workplace dangers, minimizes human error, and saves labor expenses. The KSEC Electronic Belt Scale Model K62E is designed for accurate and continuous weighing of bulk materials on conveyor belts. A prairie dog-optimized multi-kernel support vector machine (PDO-MSVM) is proposed for fault detection in conveyor systems. Several types of faults, including ball bearings, central shafts, pulleys, idler roller faults, belt slippage and drive motors, are common in belt conveyor systems. Data collected from the conveyor system, including operational logs and fault records. To analyze the obtained data, the data was first preprocessed using normalization. Features are obtained by applying a short-time Fourier transform (STFT) on preprocessed data. It may be the case that not all extracted features are relevant to the classification of faults. Therefore, in the context of the PDO application, only the most important features are selected and used for classification. The proposed method is implemented using Python. The performance of the proposed method is assessed employing measurement criteria of precision, where the proposed method attains 92%. For this case, the result demonstrated the efficiency of the proposed method compared to the other common algorithms in the fault diagnosis of the conveyor system.</p>