RES2020-04: A Localized Safety Performance Functions (SPFs) Approach Accounting for “Within” Tennessee Variations on
Freeways & Interchanges
| 项目名称： | RES2020-04: A Localized Safety Performance Functions (SPFs) Approach Accounting for “Within” Tennessee Variations on Freeways & Interchanges |
| 摘要： | Higher speeds and design features such as narrow and sharply curved ramps can hinder the safety performance of Freeways and Interchanges. It is important for practicing engineers to quantify the safety performance of such facilities so they can take appropriate measures during the planning, design, construction, operation, and maintenance. The Highway Safety Manual (HSM) is a vital resource for making safety decisions. The Tennessee Department of Transportation (TDOT) is planning to use crash predictive models in the HSM to quantify safety on its roadways. Factors such as terrain, climate, animal population, driver population, and post-crash evaluation protocols may vary across different jurisdictions/states. Hence, the HSM highly recommends calibrating the SPFs using local data or developing jurisdiction-specific SPFs, consistent with the TDOT HSM project. While state-specific SPFs can represent local conditions in Tennessee better than uncalibrated or calibrated SPFs, developed for freeways and interchanges by Bonneson et al. (2012), they are still variants of the “one size fits all” approach. They do not fully account for the uneven distributions of crashes because traffic crash frequencies and associated factors (e.g., traffic volumes) can still vary significantly and non-linearly across similar road geometries and environmental conditions within a jurisdiction. For example, the nature of crash occurrences in Knoxville and Chattanooga can be entirely different due to spatial and temporal heterogeneity. The association between crashes and key factors is often heterogeneous. In modeled relationships, it is important to correct for heterogeneity that arises from a number of observed and unobserved factors relating to (but not limited to) driving behaviors, vehicle types, socioeconomic factors, and road geometrics. Because of heterogeneity, it is important that data from Chattanooga not be used to understand and predict crashes in Knoxville, and vice versa. Recently, geo-spatial crash data, new analysis methods and computational capabilities that provide opportunities to develop “highly localized” SPFs have become available. These customized SPFs are more accurate in crash prediction on specific segments of Freeways and Interchanges. Our research team has recently introduced, in national forums, “localized safety performance functions” that can help state DOTs make more accurate crash predictions while |
| 资金： | $174,966.00 |
| 资助组织： | Tennessee Department of Transportation |
| 项目负责人： | Jonas-Fields, Stephanie |
| 执行机构： | University of Tennessee, Knoxville |
| 开始时间： | 20191101 |
| 预计完成日期： | 20211130 |
| 主题领域： | Design；Planning and Forecasting；Research |
相关文献
检索历史
应用推荐
