A Comprehensive Probabilistic Model of Traffic Loads based on Weigh-in-Motion Data for Applications to Bridge Structures

Abstract

In various efforts to assure safety and serviceability of a bridge structure throughout its lifetime, it is essential to accurately estimate the traffic load effects. Although traffic loads involve large uncertainties and can vary significantly with site-specific traffic environments, bridge design codes and maintenance strategies do not utilize a probabilistic model that can reflect the actual environments and uncertainties of the target bridge. Rapid developments of weigh-in-motion (WIM) technologies now make it possible to collect various types of data describing the characteristics of vehicles and traffic patterns. Based on actual WIM data collected in South Korea, this paper develops a comprehensive probabilistic model describing the characteristics of vehicles and traffic flow so that the traffic load effects of a target bridge can be assessed using a Monte Carlo simulation approach. To describe the characteristics of vehicles and traffic flow in the WIM data, several important random variables are first identified. These key random variables are then incorporated into a comprehensive probabilistic model based on fitted probability distributions and theories of transportation engineering. The developed model is successfully verified by comparing the daily maximum total loads estimated using actual WIM data with those estimated using artificial WIM data generated from the model. Furthermore, bridge traffic load effects, e.g., moment and tension, are estimated using the influence lines of an actual cable-stayed bridge in South Korea (the Incheon Bridge) and are compared with those from the live load model of a design code. Finally, a brief parametric study is performed to explore the possibility that a probabilistic model developed by the proposed approach can be used as a generic probabilistic traffic load model capable of estimating the site-specific traffic loads through customizations based on partial measurements and available information regarding the target bridge.