Traffic volumes are an essential input to many highway planning and design models; however, collecting this data for all road network segments is neither practical nor cost-effective. Accordingly, transportation agencies must find ways to leverage limited ground truth volume data to obtain reasonable estimates at scale on the statewide network. This paper aims to investigate the impact of selecting a subset of available automatic traffic recorders (ATRs) (i.e., the ground truth volume data source) and incorporating their data as explanatory variables into a previously developed machine learning regression model for estimating hourly traffic volumes. The study introduces a handful of strategies for selecting this subset of ATRs and walks through the process of choosing them and training models using their data as additional inputs using the New Hampshire road network as a case study. The results reveal that the overall performance of the artificial neural network (ANN) machine learning model improves with the additional inputs of selected ATRs. However, this improvement is more significant if the ATRs are selected based on their spatial distribution over the traffic message channel (TMC) network. For instance, selecting eight ATR stations according to the TMC coverage-based strategy and training the ANN with their inputs leads to average relative reductions of 35.39% and 13.44% in the mean absolute percentage error (MAPE) and error to maximum flow ratio (EMFR), respectively. The results achieved by this study can be further expanded to create a practical strategy for optimizing the number and location of ATRs through transportation networks in a state.
This research aims to study application of support vector machine algorithm, artificial neural networks and five different types of decision trees in predicting mode choice of freight transportation. Performance of these models has been compared with log it model which is one the most prevalent statistical models in the field. Effect of factors such as cargo weight, distance, type and characteristics of commodity has been studied in process of modelling mode choice which is rail and road. In this regard, data gathered in the United States, is used and similarities and advantages of the models are described in details. Results indicated that cost-sensitive support vector machine is the best method in predicting shipment mode choice. After this method, stand C5 decision tree and artificial neural network. The most important variables in determining shipment mode choice of firms are respectively weight, great-circle distance between origin and destination, commodity type, compound impedance factor of rail and truck and containerized condition of the shipment to be moved.
This paper introduces an innovative way to incentivize increased person throughput on priced highway facilities such as toll roads, express toll lanes, and high-occupancy toll (HOT) lanes, using cash rewards for carpoolers and transit riders. An exploratory evaluation of the concept is demonstrated for the I-66 HOT facility inside the Capital Beltway in Northern Virginia. The results of the analysis suggest that the concept could provide significant benefits at a relatively low cost to public agencies for cash incentives.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.