Forecasting Travel Time

Abstract

If travel time and speed are to be used as critical performance measures in congestion management systems and air quality analysis procedures, existing modeling techniques will need to be enhanced. Many of the simplifying assumptions that are built into traditional modeling techniques are described. Several relatively simple enhancements to existing models that can greatly improve the model's ability to estimate travel time and speeds are identified, and more advanced methods that could be considered as part of major model redevelopment efforts or detailed air quality studies are suggested. One of these methods involves simulation techniques. The problems and issues of integrating simulation models with travel demand forecasting techniques are outlined, and it is concluded that modeling speed is considerably more difficult than modeling volumes. The bottom-line criterion for any model enhancement is that the procedure supports decision makers in a timely and cost-effective way. This criterion is likely to limit the types of enhancements that are possible.