Procedure to estimate missing year-to-year traffic counts at intersections
Conference Paper
Overview
Identity
Additional Document Info
View All
Overview
abstract
In order to estimate accurately the safety at intersections or at any other facility, it is important to have available accident and traffic flow data for as many years as possible. Indeed, better and more robust accident prediction models (APMs) can be developed with a larger sample size. In some circumstances such as when one intends to create APMs that incorporate trend, it is necessary to have a complete database for the entire study period. Unfortunately, since manual traffic counts are very expensive to perform and the financial resources of transportation agencies are quite limited, it is impossible to record traffic at every intersection for every year. To overcome this drawback, it is necessary to find a method that enables an agency to estimate traffic counts for the years and time of day that they were not performed. Thus, the objective of this paper is to describe a procedure that was used to estimate the missing traffic counts for 1,541 signalized and unsignalized intersections in Toronto, Canada. The procedure is divided into four sequential steps and is used to explain: 1) how to expand 8-hour counts to the annual average daily traffic (AADT) counts; 2) how to estimate the missing AADTs; 3) how to compute the flows for each leg and movement (i.e, left, right, through); and 4), how to predict hourly flows for different time periods. An example is presented for each step to better illustrate the procedure. The proposed method proved to be adequate to estimate missing traffic flows at intersections.
