Incorporating weather: a comparative analysis of Average Annual Daily Bicyclist estimation methods

摘要

Average Annual Daily Bicyclists (AADB) is commonly used in a wide range of cycling-relatedresearch and practical applications. It is generally estimated by averaging the daily cyclist totalsrecorded by a long-term automatic counter, or by using such a counter to extrapolate short-termcounts. The latter method is commonly referred to as the expansion factor method, and has beenshown to produce estimates with considerable error. To help mitigate this error, this studyproposes two AADB estimation methods, one of which uses a cycling-weather model to adjustshort-term counts, and one of which is based on individual daily totals from a long-term countsite (as opposed to annual averages by day or by month). These methods are compared to twomore traditional expansion factor methods. The weather and disaggregate methods out performedthe traditional methods, with the latter producing an average absolute relative error ofroughly 14% when based on just one day of short-term data.