Traffic volumes determine both the number of travelers who will benefit from a highway improvement project and, in the case of capacity enhancement projects, the future congestion relief provided by the project. Accordingly, accurate forecasts of traffic volumes are critical to obtaining valid results from BCA.
Traffic forecasting is often more complicated than it first appears. An assumption that the historic growth rate of traffic on a road will continue unchanged after it is improved can lead to significant miscalculations of its actual future traffic. In fact, traffic levels on an improved road may increase faster than anticipated as drivers seek to take advantage of its better driving conditions.
Why would a road attract more traffic volume after it is improved than it would if it were not improved?
Drivers who formerly avoided the facility because it was too congested may start to use it once the congestion has been reduced by an improvement. Many of these drivers will divert to the improved facility from other congested regional roads. Similarly, some drivers who formerly traveled in off-peak hours on the facility to avoid severe congestion will shift back to peak hours, adding to peak hour volumes when congestion is most noticeable to commuters.
Other drivers will "unchain" existing trips into multiple trips or make new trips that they might otherwise have avoided due to excessive delay associated with congestion. Some individuals may shift from transit to automobile. Drivers may also make longer trips (to more remote locations) than they did before the improvement. Other traffic responses can and do occur.
The new and diverted users of the improved facility will enjoy benefits, just as will the existing users. These additional users, however, will use up some of the capacity of the improved facility, reducing the congestion relief that would have resulted for existing users had the additional users not arrived.
The traffic forecasting process begins with the collection of data on current traffic on the facility and throughout the region, followed by the calculation of expected growth in traffic for the region in general. This base case regional traffic projection should reflect expected economic, demographic, and land use trends, based on historic and projected relationships between these factors and regional traffic growth.
Data on expected regional traffic growth can then be entered into the region's travel demand model to simulate regional traffic flows with and without the new highway capacity. MPOs and States typically maintain the travel demand models for planning purposes. Most travel demand models now in use are effective at measuring the extent to which existing network traffic will divert to new capacity - a major source of "new" traffic on improved roads. Other traffic responses can be approximated even when they are not measured explicitly by the models. For instance, the models can be manipulated, through various feedback adjustments, to simulate the effects of mode shifts and alternative destinations chosen by regional travelers in response to a reduction in congestion. Although not explicitly captured in most travel demand models, the shifting of traffic to and from peak periods as congestion levels change can be estimated using supplemental methods.
A travel demand model may indicate that a significant amount of future traffic on the facility to be improved will be diverted from other roads in the region. This effect, while mitigating some of the congestion relief on the improved facility itself, will reduce congestion on the other roads. In this case, the BCA for the new capacity project should attempt to incorporate the beneficial effects (as measured by the travel demand model) of the improved facility on other roads in the corridor as well as on the facility itself. Of course, reduced delay on the affected roads may lead to some compensating, new trip generation on those roads as well.
Unless the analyst considers traffic responses to an improved facility, he or she may overestimate the benefits of the improved facility to existing users and understate the benefits to the new users and those drivers on other roads in the regional highway network. This can lead to misperceptions by decision makers and members of the public about the important benefits of new capacity on regional traffic patterns and congestion.
On the improved facility, the time saving benefit per trip for pre-existing users will diminish relative to what it would have been had traffic volume not changed, but will still be positive. This saving can be calculated directly from the reduction in delay based on changes in the volume/capacity ratios caused by the improvement, after allowing for traffic adjustments. Users of other routes in the network who do not divert to the improved facility will similarly receive time saving benefits caused by the reduction in traffic volume due to the diversion of others to the improved facility.
Users on the improved facility who diverted from other routes will receive benefits equal, on average, to the midpoint between those of pre-existing users of the improved facility and those of users of other facilities who do not divert from those facilities. This midpoint value reflects the fact that some diverted users will gain the full time saving of the improved facility but others will do only slightly better than had they not diverted. By a similar computation, users making new (as opposed to diverted) trips on the improved facility or other routes can be shown to experience benefits equal, on average, to half of those experienced by pre-existing users on the respective facilities.
Numbers of affected users for each user class, along with data on the amount of time saving, can be derived from the travel demand modeling procedures described in this section.
Standard travel demand modeling, principally addressing trip diversion, is often sufficient for BCA of routine capacity projects. State or MPO planning offices often undertake such modeling as a matter of course in their preparation of transportation improvement plans. In general, it is a good idea to conduct BCA in close coordination with planning offices.
A comprehensive traffic forecast, incorporating the full range of traffic responses to capacity improvements, should be done for regionally significant or controversial projects. It is easy for the credibility of the BCA to be challenged if it is learned that new trips or other effects of new capacity were ignored. Traffic forecasting can be used to educate the public that the new capacity leads to benefits for more than just the existing users of the to-be-improved road, and that traffic diverting to the new road will reduce congestion throughout the network.
Price Elasticity of Demand and Traffic Forecasting
An important benefit of a capacity expansion project is the reduction in travel times for highway users. Travel time is a major component in overall price or cost to the user, which includes time as well as out-of-pocket costs. As with most goods and services, a lower price can be expected to lead to more quantity demanded - in this case, some additional travel.
Price elasticity of demand is an economic concept used to summarize how much more or less of something people will consume if its price changes. From the standpoint of estimating future traffic levels, elasticity represents how a change in the cost of driving, due to a reduction in travel time or implementation of a toll, may affect the volume of travel that will take place. These changes in volume result from some drivers' decisions to make more or fewer trips than they otherwise would have made.
Elasticity is stated in percentage change terms, e.g., an X percent reduction in travel price leads to a Y percent increase in travel miles or trips. An elasticity of zero implies that travel is unresponsive to a price change, no matter how large, while an infinite elasticity implies that even a one-second decrease in travel time will cause all capacity to be completely absorbed.
While price elasticity is a generally accepted tool in economics, there are differing opinions about how to apply it in a transportation context. The transportation economics literature reveals a wide range of measured elasticity values, reflecting different study methods, data, time periods, and locations. No studies, however, suggest that travel demand elasticity is either zero or infinite. When measured on a given facility, observed elasticity includes the effects of both diverted trips, which represent existing traffic that has simply shifted from other routes or time periods, and new travel taken as a consequence of the lower user cost. Additional research is needed to narrow the range of elasticity values that are applicable to a given set of circumstances - whether facility, corridor, or region - and to develop methods for better incorporating demand elasticity into traffic forecasting.