Gao, Dongjian. Reliability-based assessment of load and resistance factor rating for specialized hauling vehicles. Retrieved from https://doi.org/doi:10.7282/t3-kff7-wg83
DescriptionSpecialized Hauling Vehicles (SHVs) are defined as short heavy trucks within the legal weight limits but induce higher load effects than routine commercial load models.
This thesis is consisted of three major parts, 1) investigation of the population, various truck characteristics of SHV traffic fleet and multiple-presence events based on available Weigh-In-Motion (WIM) data, 2) development of statistical models of SHVs for both one-lane and two-lane loading cases, and 3) assessment of live load factors for SHVs for LRFR Strength I Limit State.
The first part focuses on processing WIM data. The algorithm for data processing and the WIM database is presented first. The database consists of two sets of data: 1) data from New Jersey, and 2) data from another 21 states across the United States. Based on the data processing results, the statistics of SHV traffic in terms of percentage of entire truck traffic and average daily count at each WIM site were obtained. Also, the frequency distributions of number of axles, gross vehicle weight (GVW) as well as a summary of typical SHV configurations were acquired by assembling the SHV traffic from all sites. The multiple-presence events were also assessed, and the statistics are presented. The second part presents the analysis of load effects of SHVs and the development of statistical models for SHVs. The statistical models consist of three factors: bias ratio, coefficient of variation and probability distribution. Due to the multiple-presence events of SHVs, the models were developed for both one-lane loading case and two-lane loading case. The probability distribution of load effects induced by SHVs on various spans was determined by hypothesis testing. Based on the probability distribution, the maximum load effects over a 5-year period of bridge evaluation were predicted for each site. The site maximum values were then assembled into one dataset and the mean and coefficient of variation were calculated as the statistical parameters for live load. The probability distribution of the site maximum values was also determined by hypothesis testing. In the third part, the statistical models for live loads were used for reliability analysis for existing steel-girder bridges. The load rating factors were obtained for the bridges and plotted against reliability indices for different combinations of SU trucks and ADTTs. The results show that the reliability level provided by the current live load factors are higher than the desired level. The live load factors as well as the live load factors were modified until reliability index equal to 2.5 when rating factor equal to 1 in each plot. New live load factors for SHVs for different ADTTs were presented based on the assessment results.