Acoustical characteristic changes due to transverse rumble strips installation on roadway

Abstract

Transverse rumble strips (TRS) acts as a traffic calming device to alert inattentive drivers through the optical, sound and vibration effects. Currently, the existing TRS guidelines have been found to be very basic with no detailed explanation on the TRS profile. Inappropriate design of TRS profile may generate excessive external noise level that can affect the nearby residents. Besides, the real cause of TRS roadside noise annoyance is still unknown. Therefore, this study aims to evaluate the acoustical characteristic changes due to TRS installation at residential areas in contributing to the framework of optimum TRS design selection for future installation. The objectives of this study are to investigate current TRS designs and characteristics, determine road and traffic characteristics at selected roadways, evaluate noise annoyance due to TRS installation and develop empirical model of TRS roadside noise level for the proposal of framework of optimum TRS design selection. Visual investigation and on-site measurement were conducted to investigate the current TRS design and characteristic. Measurement of road and traffic measurements were also carried out to evaluate the skid number (SN), vehicle speed, traffic volume and composition that may affect sound level changes. This study measured the external noise generated by a range of gross vehicle weight (GVW) between 850 kg and 7700 kg which travel over three types of TRS, namely middle overlapped (MO), middle layer overlapped (MLO) and raised rumbler (RR) with speed range of between 30 km/h to 70 km/h. Controlled pass-by (CPB) method was conducted to determine the sound level changes. Actual traffic condition due to TRS installation was also measured. Meanwhile, assessment of sound level changes and impulse, low frequency noise (LFN) and tonal, noise limit and traffic noise index (TNI) were carried out to assess the noise annoyance from TRS. Statistical analyses were used to test all related variables of roadside noise level due to TRS. Significant variables were used for the development of empirical noise models by using multiple linear regression. Then, framework of optimum TRS selection design was proposed to assist local authorities for future TRS installation. TRS thickness was limited to 3 mm for all profile types. TRS roadside noise may affect nearby residents as the sound level increment was more than 6 dBA. The sound level changes for vehicles transit on MLO were higher than MO and RR of more and less than 3 dBA, respectively. Besides, MLO showed the most significant impulsivity regardless of the multiple vehicles transit. TRS roadside noise was also characterized by LFN at frequency region of 20 Hz to 125 Hz. Tonality was also found significant especially due to heavier vehicles with GVW of 5000 kg and above at speed of 50 km/h. Among the three types of TRS, MO showed the best performance due to its insignificant acoustical characteristics. SN, vehicle speed, traffic volume and GVW were found to be significant variable for the generation of TRS roadside noise level. Meanwhile, other variables, such as TRS width, length, spacing, profile and strip number, traffic volume and skid resistance were found insignificant. The established empirical models based on the real causes of TRS noise annoyance were used in the proposal for framework of optimum TRS design selection for future installation. This proposed framework could enable exploration of TRS impact on the traffic noise condition to facilitate the local authority on the best selection of TRS type during the surround area become critical for noise annoyance.