This report describes how Replica model compares to other data sources describing movement in the region, including relevant Census surveys aggregates and observed volumes of traffic, transit boardings and non-motorized trip counts in the region. Replica describes a typical week of travel for a specific three-month (quarter) period, and comparisons are made with the data collected within the relevant period and aggregated to represent a typical day of travel.¶
Each section of the report provides details on the accuracy of the model within the relevant comparison category.¶
This section compares overall county-to-county work commute characteristics against Census Transportation Planning Product (CTPP) data. Replica models all trips by residents of the core modeling region as well as trips by residents of the extended region who work in the core. A comparison of the distribution of number of departures within each hour by county of origin is provided.¶
Note: Replica counts are scaled so that the total number of trips matches the CTPP, in order to account for discrepancies between average daily travel patterns and self-reported survey responses.¶
CTPP movements, Figure 1: Commute departures by hour by county¶
CTPP 2012-2016 is used as the source of commute volumes for county-to-county and significant tract-to-tract flows, as well as reported departure times.
Commute flows should only be compared with CTPP for typical weekdays (not weekends).
Note: Error and Bound numbers for this section are 1 - p, where p is the correlation coefficient.¶
Passenger vehicle volumes are compared to averaged seasonal daily volumes collected at multiple locations within the region and aggregated over different road volume groups and road categories as recommended by FHWA.¶
Auto, Figure 1: Map of the traffic volumes sensor locations for the simulated region/season.¶
Total hourly volumes are generally less accurate in the night hours. Due to privacy considerations Replica does not reproduce most of the rare nighttime (not “typical”) trips.¶