Toward a better estimation of annual average daily bicycle traffic: Comparison of methods for calculating daily adjustment factors

Abstract

Several methods exist to calculate and apply adjustment factors for bicycle traffic. The reported accuracy of these factors differs from one study to another, making it difficult for transportation professionals to decide which method is preferable. The objective of this study is to compare the estimation accuracy of methods used to calculate daily adjustment factors and to quantify the performance of these methods relative to their mathematical intensity. Three methods of calculating daily adjustment factors were studied: the AASHTO method, the monthly and weather-specific method, and the day-of-year method. The estimation accuracy of annual average daily bicycle traffic volumes was assessed. The results supported the superiority of day-of-year factors, for which the mean absolute percent error (MAPE) was about 17.5%. Second was the monthly and weather-specific factors (MAPE ≈ 24.5%), and third was the AASHTO factors (MAPE = 30.0%). Detailed error analysis was carried out to select the best days for bicycle volume data collection. The daily factor was modeled against day-specific attributes, such as weather conditions and day type, in an attempt to compute generic factors that are applicable to other locations. The model showed very good fit to the data with a coefficient of determination of about.87. The model was further used to develop daily factors for the validation data set and hence calculate annual average daily bicycle volumes. The MAPE was found to be 26.4% for all days and 20.5% for weekday data. The paper provides insights on the advantages and disadvantages of each calculation method and shows the priority preference of application.