Developing a method for estimating AADT on all Louisiana roads.

Details

• Creators:
  Sun, Xiaoduan ; Das, Subasish
• Corporate Creators:
  University of Louisiana at Lafayette. Department of Civil and Environmental Engineering
• Corporate Contributors:
  United States. Federal Highway Administration
• Subject/TRT Terms:
  • Annual average daily traffic
  • Traffic counts
  • Traffic estimation
  • Traffic flow
  • Traffic volume
  • Rural highways
  • Sensitivity analysis
  • Regression analysis
  • Pattern recognition systems
• Publication/ Report Number:
  FHWA/LA.14/548
• Resource Type:
  Tech Report
• Geographical Coverage:
  Louisiana
• Corporate Publisher:
  Louisiana Transportation Research Center
• Abstract:
  Traffic flow volumes present key information needed for making transportation engineering and planning decisions. ; Accurate traffic volume count has many applications including: roadway planning, design, air quality compliance, travel ; model validation, and administrative purposes. Traffic counts also serve as an important input in highway safety ; performance evaluation. However, collecting traffic volume on all rural non-state roads has been very limited for various ; reasons, although these roads constitute a great portion (60 to 70%) of road mileage in the roadway network of any state in ; the U.S. For example, out of 61,335 miles of roadway in Louisiana, 73% of the roadways are non-state roads. Due to ; limited resources, traffic volume information on non-state roadways has not been systematically collected in Louisiana. ; Generally, traffic volumes on these roads are fairly low, and VMT on these roads is much less compared with that on ; interstate or arterial roads. Thus, regularly conducting traffic count is not economically feasible for non-state roadways. This ; study develops an AADT estimation methodology by using modern statistical and pattern recognition methods. By using ; available traffic counts on non-state roadway and four variables (namely: population, job, and distance to intersection and to ; major state highways at block level), a training set to estimate roadway AADT for eight parishes were obtained by a ; modified support vector regression (SVR) method. This pattern recognition method yields better AADT estimates than the ; conventional parametric statistical methods. Sensitivity analyses were also conducted in this study, which indicates a parish-specific model works better than an aggregated single model. ; With the estimated AADT, the DOTD and local government agencies can make better decisions on funding allocations for ; safety improvement projects and pavement maintenance actions. The estimated parish-specific AADT on non-state roads ; can also improve statewide travel demand forecasting models and air quality assessment.
  More +
• Format:
  PDF
• Funding:
  LTRC No. 14-3SA;SIO No. 30001700
• Download URL:
  https://rosap.ntl.bts.gov/view/dot/29681/dot_29681_DS1.pdf
• File Type:
  [PDF - 3.80 MB]
• Collection(s):
  US Transportation Collection
• Main Document Checksum:
  urn:sha-512:610150ab82e78e0b60f7e3dc2cc3171f80dad37531f3ffaa47d8008fbeb6b129e1f2f6308022fd1ac53de31db38b6b3a2c0304dd0df194a1d4288a212fbee0d2