Post-construction evaluation of traffic forecast accuracy

The analysis indicated that the impact of new rail projects on increasing total transit ridership was minimal and actual ridership in most of the cities considered fell far short of the ridership forecasts available to the decision-maker at the time of deciding to go ahead with the project. [...] The estimation of average inaccuracy shows that the average inaccuracy is less than zero in 47% of the projects and the average inaccuracy is greater than zero in 49% of the projects. [...] The basic functional form of the regression model estimated is I¼ f ðN,H,F,V ,D,T ,SÞ ð2Þ where I is the Inaccuracy ratio estimated as the difference between the forecast and actual traffic, standardized by the actual traffic, N the Number of years between report year and forecast year, H the Roadway type, F the Functional Classification (used in Model 1 alone), V the Project size measured in VKT,. [...] The view of the interviewees was that criticisms against the use of modeling in forecasting arises when, for example, results are used by policy makers who lack an under- standing of the process behind the numbers or when policy makers apply a macro level model to a micro level study area without adequate changes to the parameters of the model given the difference in the scale of analysis. [...] The results from the comparative analysis indicate the presence of errors in the demographic forecasts used in the travel demand models, which may have contributed to the inaccuracy in the roadway forecasts.