TRAFFIC FORECASTING IN PRACTICE

With the basic procedures outlined in the previous sections of this chapter, a traffic forecasting model similar to those used in practice can be developed and implemented. Although there are many subtleties in the process that are beyond the scope of this book, the basic procedure is as follows

1. The geographic region being studied is segmented into nearly homogeneous areas based on similarities in land use,

socioeconomic conditions, and so on. These areas are referred to as traffic analysis zones (TAZs) and are used to determine the origins and destinations of trips. The choice of the number of TAZs (often simply referred to as zones) presents a trade-off between accuracy (smaller TAZs provide more detailed forecasts) and ease of implementation (larger TAZs require less data and are easier to incorporate in the overall model system). A single point is usually chosen within the TAZ as the assumed origin/destination point of all TAZ trips. This point is referred to as the centroid.

2. The highway network is defined to include the relevant highway segments. Highway segments are linked by using nodes, which are usually placed at intersections or other points where highway capacity could change. Nodes permit traffic to travel from one highway segment to the next. The highway network is a representation of the actual street network and carries traffic flow between TAZs. As was the case with the size of the TAZs, a very large and detailed highway network can provide very detailed forecasts but also requires large amounts of data, and thus smaller networks are often used. Defining the highway network includes detailed information on each highway segment’s performance function so that traffic flows can be computed (usually by assuming user-equilibrium route choice). The performance function often used in practice to relate traffic flow with travel time was originally developed by the U.S. Bureau of Public Roads and takes the form:

[1 ( / ) ]n fn n cnt t x x βα= + (8.10)

where

tn = travel time on highway segment (route) n, usually in minutes, tfn = free-flow travel time on highway segment (route) n, usually in minutes, xn = traffic flow on highway segment (route) n, usually in veh/h,

xcn = capacity of highway segment (route) n, usually in veh/h, and

α, β = model parameters that usually vary with respect to the capacity and speed limit of the highway segment (route).