If you want to estimate Annual Average Daily Bicyclists/Pedestrians (AADBP) at locations where less than 12 months of count data is available, expansion factors are a helpful way to do it. These adjust short duration counts to the average. For example, if a short duration count is collected in June, expansion factors allow us to account for the fact that a day in June might be expected to have higher volumes than an average day. Compute these factors annually, if possible.
There is a growing literature on how best to compute expansion factors for bicyclists and pedestrians. Some methods are based on similar methods used for motor vehicles; other methods include weather variables; and still other methods compute separate factors for each day of the week. Here we offer a list of resources to help you compute factors for a single permanent counter or a group of permanent counters.
Alex Hyde-Wright has created an example of a simple method for estimating factors for one permanent count station. This example can be downloaded here.
A traditional method for computing factors is documented in this PDF.
Similar methods are described in the Traffic Monitoring Guide.
If no permanent bicycle or pedestrian count data are available in your region or state, the National Bicycle and Pedestrian Documentation Project, a joint effort by Alta Planning and Design and the Institute of Transportation Engineers, provides a set of factors that can be downloaded from their website (see the Extrapolation Workbook). Since bicycle and pedestrian traffic patterns vary greatly by geography and climate, applying these national factors can result in large error and may only be appropriate for very rough estimates.
El Esawey and others, working with data from Vancouver, British Columbia, have investigated the details of how to estimate hourly, daily and monthly factors, including investigating how to include weather factors. Their first paper, available here, discusses the best approaches to computing daily factors specifically. Their second paper analyzes both daily and monthly factors and can be found here in addition to their TRB presentation.
Others have also investigated factoring including how weather might be included:
- Sears, Flynn, Autlman-Hall and Dana 2012
- Nordback 2012
- Dowds and Sullivan 2012
- Figliozzi, Johnson, Monsere, Nordback 2014
Nordback and others have created a set of Colorado-specific factors published in their 2013 report.
Recent work presented at the 2014 annual meeting of the Transportation Research Board by two independent teams, Hankey, Lindsey and Marshall in Minneapolis1, and Nosal, Miranda-Moreno and Krstulic in Montreal2, found that using factors for each day of the year could outperform traditional seasonal adjustment factors even when weather was specifically factored in. The Montreal study is available here.
4.4.6 STEP 7: COMPUTE ADJUSTMENT FACTORS
The computation of adjustment factors should follow a similar process as motorized traffic volumes outlined in Section 3.2.1. These adjustment factors will be calculated for each unique non-motorized traffic factor group as determined in Step 4.
In practice, very few agencies have applied monthly or DOW adjustment factors to short-duration non-motorized counts. The current prevailing practice is to collect short-duration counts during those dates and times that are believed to be average, thereby reducing the perceived need for adjustment. However, this practice should evolve to a more traditional traffic monitoring approach as more permanent non-motorized count locations are installed.
1Hankey, S., G. Lindsey, et al. (2014). Day-of-Year Scaling Factors and Design Considerations for Non-motorized Traffic Monitoring Programs. 93rd Annual Meeting of the Transportation Research Board. Washington, D.C., National Academies.
2Nosal, T., L. Miranda-Moreno, et al. (2014). Incorporating weather: a comparative analysis of Average Annual Daily Bicyclist estimation methods. 93rd Annual Meeting of the Transportation Research Board. Washington, D.C., National Academies.