Tractors are essential for many farming tasks but cause high vibrations that lead to operator discomfort and fatigue. This study examines how different tractor settings during tillage operations affect Hand-Arm Vibration (HAV). The settings tested were speed (0.6, 0.7, 0.8 m/s), draft setting (2, 4, 6 kN), and tillage depth (0.10, 0.12, 0.14 m), following ISO 5349-1:2004 standards. The Taguchi L27 array for designing the experiments, Response Surface Methodology (RSM) to see how different settings affect the results of HAV responses along the x, y, and z axes for each experiment. Experiments showed that vibrations were highest along the z-axis. Rotavation caused more HAV than harrowing and cultivation. As speed increased, daily HAV exposure also rose significantly. Analysis showed that speed and draft setting had a major impact on HAV levels. The study used different models to predict HAV, finding the quadratic model to be the most accurate. Optimal settings to minimize HAV were a speed of 0.8 m/s, draft setting of 2 kN, and tillage depth of 0.14 m. An artificial neural network (ANN) model also predicted HAV accurately with just a 2 % error. The findings suggest that the ANN model effectively predicts HAV under various tractor settings with constrain to the selected input setting. Relevance to the Industry: This research highlights the measures to reduce hand-transmitted vibration by optimizing the input (riding) parameters among tractor operators, which offers to improve the health and safety of users and reduce fatigue in actual farm conditions. In addition, the ANN model helps predict the HAV response under different input (riding) conditions. Ultimately, it is beneficial for the manufacturers and agriculture practitioners to optimize the tractor design and usage, ensuring safer and more efficient farm activities.

来源平台：COMPUTERS AND ELECTRONICS IN AGRICULTURE