The changing accuracy of traffic forecasts

Abstract

Researchers have improved travel demand forecasting methods in recent decades but invested relatively little to understand their accuracy. A major barrier has been the lack of necessary data. We compiled the largest known database of traffic forecast accuracy, composed of forecast traffic, post-opening counts and project attributes for 1291 road projects in the United States and Europe. We compared measured versus forecast traffic and identified the factors associated with accuracy. We found measured traffic is on average 6% lower than forecast volumes, with a mean absolute deviation of 17% from the forecast. Higher volume roads, higher functional classes, shorter time spans, and the use of travel models all improved accuracy. Unemployment rates also affected accuracy-traffic would be 1% greater than forecast on average, rather than 6% lower, if we adjust for higher unemployment during the post-recession years (2008 to 2014). Forecast accuracy was not consistent over time: more recent forecasts were more accurate, and the mean deviation changed direction. Traffic on projects that opened from the 1980s through early 2000s was higher on average than forecast, while traffic on more recent projects was lower on average than forecast. This research provides insight into the degree of confidence that planners and policy makers can expect from traffic forecasts and suggests that we should view forecasts as a range of possible outcomes rather than a single expected outcome.

Measuring and Modelling the Concentration of Vehicle-Related PM2.5 and PM10 Emissions Based on Neural Networks

“…Traffic-related emissions are primarily determined by the main characteristics of the traffic flow, such as the number of vehicles, speed, and standing time at the intersection [51,52]. Building prediction models on a dataset of many variables with relatively few observations can provide less accurate predictions and limit the performance of deep learning models [53].…”

Access-based land value appreciation for assessing project benefits

“…First, in accordance with numerous retrospective transport project evaluations, the ex ante estimations on project costs and benefits based on this approach have been repeatedly reported to be highly inaccurate (Andri et al, 2019, Cantarelli et al, 2012, Cruz and Sarmento, 2019, Flyvbjerg et al, 2004, 2005, Hartgen, 2013, Hoque et al, 2021, Huo et al, 2018, Lee, 2008, Li and Hensher, 2010a, Love et al, 2016, Lundberg et al, 2011, Nicolaisen and Driscoll, 2014, Odeck, 2004, Park and Papadopoulou, 2012, Parthasarathi and Levinson, 2010, Pickrell, 1989, Sebastian, 2005, Voulgaris, 2019a,b, Welde and Odeck, 2011, Wang and Levinson, 2023a. The estimates and verification of project costs are more straightforward and comparable among multiple projects than those of economic benefits.…”

Short-Term Traffic Flow Forecasting A wide and deep approach with periodic feature selection

“…The goal of short-term traffic forecasting is to provide an accurate prediction of the future traffic states within a time horizon ranging from 5 min to 60 min [4], [5]. The accuracy of current prediction has been found in [6] to be on average 17% in terms of Mean Absolute Percentage Error (MAPE) between the projected forecasts and the true traffic volumes based on complied traffic forecast accuracy data from 1291 projects in both Europe and the United States. Surprisingly, whilst standard in traffic modelling, GEH and other traffic modelling criteria are not used to evaluate forecasting model performance [4], [5].…”