Predicting Freeway Work Zone Delays and Costs with a Hybrid Machine-Learning Model

Abstract

A hybrid machine-learning model, integrating an artificial neural network (ANN) and a support vector machine (SVM) model, is developed to predict spatiotemporal delays, subject to road geometry, number of lane closures, and work zone duration in different periods of a day and in the days of a week. The model is very user friendly, allowing the least inputs from the users. With that the delays caused by a work zone on any location of a New Jersey freeway can be predicted. To this end, tremendous amounts of data from different sources were collected to establish the relationship between the model inputs and outputs. A comparative analysis was conducted, and results indicate that the proposed model outperforms others in terms of the least root mean square error (RMSE). The proposed hybrid model can be used to calculate contractor penalty in terms of cost overruns as well as incentive reward schedule in case of early work competition. Additionally, it can assist work zone planners in determining the best start and end times of a work zone for developing and evaluating traffic mitigation and management plans.

Survey of Work Zone Practices and Validation of Highway Capacity Manual Work Zone Capacity Model

“…Estimated pre-breakdown capacity from field-collected traffic flow parameters and HCM 6 capacity were compared using RMSE and the mean absolute percentage error (MAPE) as shown in Equations 3 and 4 respectively. Both are performance measures used in past research studies ( 7 , 23 , 24 ). The process compares the theoretical value from HCM 6 and the actual value from field data and measures the goodness of fit.…”

“…Work zones may result in lane closures or detours that often cause mobility and safety issues, arising from lane drops and merges that result in nearly 24% of non-recurring congestion, or 482 million vehicle hours of delay ( 2 , 3 ). Numerous studies abound on work zones both being the cause of significant localized congestion and raising significant safety concerns ( 4–7 ).…”

Predicting Traffic Performance During a Wildfire Using Machine Learning

“…Machine learning has also been used in previous studies for forecasting travel delay. An example of a successful application of machine learning to predict traffic delay was conducted in New Jersey ( 27 ). The goal was to determine the delay to freeway users resulting from construction lane closures.…”

Review of Methods for Estimating Construction Work Zone Capacity

“…Many researchers have adopted machine learning algorithms such as decision tree (4), ensemble learning (34,35), neural network (36-39), and support vector machine (SVM) (40) to predict the work zone capacity. The following paragraphs present an extensive review of the existing literature.…”