Methodology for Predicting MAP-21 Interstate Travel Time Reliability Measure Target in Virginia

Abstract

This study develops a target setting methodology for the (Moving Ahead for Progress) MAP-21 Interstate Travel Time Reliability Measure of “Percent of the Person-Miles Traveled on the Interstate that are Reliable” (PMTR-IS). The study uses data specific to Virginia for a set of independent variables (Hourly Volume, and Volume/Capacity Ratio, Truck Percentage, Equivalent Property Damage Only Rate, Lane Impacting Incident Rate, Number of Lanes, Presence of Safety Service Patrol, Terrain, Urban Designation) to predict whether a MAP-21 reporting segment is reliable. This is used to estimate predicted PMTR-IS with the MAP-21 specified formula. CART (classification and regression trees) models were used with 1,536 different configurations. Of the four best performing models, one was selected using engineering judgement. Datasets from 2017 and 2018 were used for training, and from 2019 and 2020 for testing. The predicted and calculated PMTR-IS were compared, and the error percentage was lower than 1% for 2019, which can be considered negligible. The 2020 error rate was higher and can perhaps be attributed to unusual reliability because of the impacts of the pandemic on travel. Sensitivity analyses revealed that the predicted PMTR-IS is reactive to capacity increase in unreliable sections at a local level, slower to respond to local AADT (annual average daily traffic) increase, and stable to small statewide AADT oscillations. The authors recommend using this methodology for target setting, taking into consideration the state authorities’ other strategic priorities.

Travel Time Reliability Prediction Using Random Forests

“…The length of TMC segment (mile), heavy vehicle percentage (heavy_percent), the number of through lanes (throu_lane), and volume-to-capacity ratio (vc_ratio) were ranked high for both GRF and QRF models, which indicates that these variables have stronger predictive power than other variables for predicting the 50th, 80th, and 90th travel times based on the study dataset. These variables are among the major influencing factors identified in current literature for modeling travel time reliability ( 7 , 26 , 36 ). Compared with all other models, the QRF interchange model gave relatively lower importance to urban/rural designation (rural), and relatively higher importance to v/c ratio, SSP and crash variables.…”

The development of BPR models in smart cities using loop detectors and license plate recognition technologies: A case study

“…Using four years of Virginia Interstate data, a modeling methodology was developed based on classification trees in Babiceanu Lahiri's work (Babiceanu & Lahiri, 2022). As independent variables, this model used the Hourly Volume, Volume/Capacity Ratio, Truck Percentage, Equivalent Property Damage Only Rate, Lane Impacting Incident Rate, Number of Lanes, Presence of Safety Service Patrol, Terrain, and Urban Designation to predict the reliability class (reliable or unreliable) of each segment in a year.…”