EVALUATION OF REDUCED TRA FFIC DATA COLLECTION PLANS USING REDUCTION-EFFECTIVENESS RATIOS

摘要

Because of budget shortfalls in recent years, state departments of transportation (DOTs) need toadjust their traffic data collection plans by reducing data collection locations and/or extendingdata collection cycles; however, only limited studies have been performed to evaluate the costeffectiveness of different data collection reduction efforts. The purpose of this study is todevelop a quantitative method for evaluating the impact of different reduced traffic datacollection plans on the overall accuracy of the annual average daily traffic (AADT) estimation.To compare the accuracy of ten reduced data collection plans with a base plan, the mean absolutepercentage error (MAPE) is calculated. In addition, the reduction effectiveness ratio (i.e., thepercent of reduced data collection cost to the percent of increased AADT estimation error) isproposed in this study. Results of this study show that while the current practice, whichrandomly selects data collection sites based on different cycles, performs well in maintainingAADT estimation accuracy, it may not be the most cost-effective approach. Results also showthat certain types of sites, e.g., rural sites, lower AADT sites, and higher AADT variation sites,tend to produce larger errors if they are not counted. These results imply that the proposedmethod provides a quantitative means to evaluate different reduced data collection plans. Itenlightens directions to enhance current data collection and traffic estimation practices. Moreimportantly, it enriches the information provided for state DOTs to make informed, andeffective, decisions under the stringent budget.