United States Department of Transportation
i
Non-Traditional Methods to Obtain Annual Average Daily Traffic (AADT)
-
2021-09-01
-
[PDF-7.64 MB]
Details:
-
Creators:
-
Corporate Creators:
-
Contributors:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Contracting Officer:
-
Corporate Publisher:
-
Abstract:The use of passive data from location-based smartphone applications (LBS) and Global Positioning Services (GPS) to collect Annual Average Daily Traffic (AADT) has the potential to greatly reduce costs to State Department of Transportations (DOTs) and Metropolitan Planning Organizations (MPOs) and expand the coverage of up-to-date counts. This report evaluates the technical and statistical validity of traffic data derived from these sources using machine learning methods. Validity was determined by comparison to 4255 permanent counters, and a survey of recent publications about accuracy expectations. The document covers the input data and the development of the machine learning models and model validation. The results include the error by road volume, roadway and regional characteristics compared to typical estimation. The effects of reduced trip sample, ping rate, spatial accuracy and reference counters were also tested. The applicability of Probe Data was tested for other factors including, day of week, month of year, directional and ramp AADT, work zones ADT, K and D factors, peak hour truck data, special events or unusual weather and AADT by vehicle type.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type:
