Iranian Society of Acoustics and Vibration Journal of Vibration and Sound 2383-1839 14 28 2026 02 20 Modeling vehicle driving parameters to optimize vibrations of cargo types in different frequency ranges using machine learning Modeling vehicle driving parameters to optimize vibrations of cargo types in different frequency ranges using machine learning 178 199 733274 FA Salman Imanpour Ph.D. Student, Faculty of Engineering, Imam Hossein University, Tehran, Iran 0009-0004-1451-7518 Saeed Mahjoub Moghadasi Associate Professor, Faculty of Engineering, Imam Hossein University, Tehran, Iran Seyed Mohammad Modarres Researcher, Faculty of Engineering, Imam Hossein University, Tehran, Iran Journal Article 2025 09 30 Vehicle vibrations can significantly impact the health of the vehicle's cargo and its overall performance. This study investigates the effects of tire pressure, vehicle cargo weight, and driving speed on vehicle vibrations using both analytical and experimental methods, aiming to optimize vibration energy in different frequency ranges and determine optimal conditions for two specific types of cargo. By analyzing the interactions between these three factors, this research seeks to identify the main factors influencing vibrations and their potential combined effects, ultimately finding vibrationally optimal states for different cargo types. Vehicle speed, tire pressure, and cargo weight were systematically varied, vibrations were measured, and the impact of each variable and their interactions on vibration levels was assessed. Three factors (speed, tire pressure, and cargo weight) were tested at three levels with four repetitions, totaling 108 experiments. The results show that vehicle speed has the greatest effect. Tire pressure and cargo weight have much smaller effects compared to speed. Also, the vibration level is highest in the vertical direction and lowest in the longitudinal direction of the vehicle. Finally, optimal parameters were calculated for two types of cargo. This was done using both traditional methods and machine learning methods. Among the machine learning methods, the decision tree algorithm, as an interpretable method, has greater accuracy than the traditional method. The results of this research have many applications in UGV, especially tactical UGV that carry sensitive cargo. Vehicle vibrations can significantly impact the health of the vehicle's cargo and its overall performance. This study investigates the effects of tire pressure, vehicle cargo weight, and driving speed on vehicle vibrations using both analytical and experimental methods, aiming to optimize vibration energy in different frequency ranges and determine optimal conditions for two specific types of cargo. By analyzing the interactions between these three factors, this research seeks to identify the main factors influencing vibrations and their potential combined effects, ultimately finding vibrationally optimal states for different cargo types. Vehicle speed, tire pressure, and cargo weight were systematically varied, vibrations were measured, and the impact of each variable and their interactions on vibration levels was assessed. Three factors (speed, tire pressure, and cargo weight) were tested at three levels with four repetitions, totaling 108 experiments. The results show that vehicle speed has the greatest effect. Tire pressure and cargo weight have much smaller effects compared to speed. Also, the vibration level is highest in the vertical direction and lowest in the longitudinal direction of the vehicle. Finally, optimal parameters were calculated for two types of cargo. This was done using both traditional methods and machine learning methods. Among the machine learning methods, the decision tree algorithm, as an interpretable method, has greater accuracy than the traditional method. The results of this research have many applications in UGV, especially tactical UGV that carry sensitive cargo. https://jvs.isav.ir/article_733274_4e5b410faf8d9732f897ac5c1d540489.pdf