Polanco-Portalatín, Máximo G.

Profile Picture

Filters

20111
Reset filters

Settings

Citations

Publication Search Results

Now showing 1 - 1 of 1
  • Thumbnail Image
    Publication
    Estimation of volume delay functions for arterials and freeways in the western region of Puerto Rico
    (2011-07) Polanco-Portalatín, Máximo G.; Valdés-Díaz, Didier M.; College of Engineering; Cruzado Vélez, Ivette; Figueroa Medina, Alberto M.; Department of Civil Engineering; Perales Perez, Oscar
    Volume Delay Functions (VDF) are formulas developed to describe the functional relationship between travel time and traffic flow in a travel demand model considering the available capacity of the facility. VDF’s are the key element of the traffic assignment process in travel demand modeling. The objective of this research was to calibrate VDF’s for Arterials and Freeways located at the Western Region of Puerto Rico. The three most common functional forms were used in the calibration process to identify the one that best describes the behavior of travel time against the volume-to-capacity ratio observed in the selected road segments. Volume and speed data were collected using vehicle counters. The vehicle counters were installed in each of the study segments. Travel time data was collected using video cameras at upstream and downstream locations on each of the segments evaluated. The VDF’s evaluated were the BPR model, the Conical model and the Akcelik model. Nonlinear regression was used to estimate the parameters of each VDF studied. In addition, the performance of each functional form is evaluated with goodness of fit measures. The results of the nonlinear regression process indicate that the best models are the Akcelik model and the BPR model. It was demonstrated that there is a significant difference on using the functions with their parameters for default versus the calibrated parameters obtained through the estimation process followed in this research, particularly under congested conditions. The mean absolute percentage errors (MAPE) of all models calibrated were compared with the MAPE of the models using the default values. This comparison showed that all the models calibrated have lower MAPE results. The BPR model calibrated for arterials produced a MAPE of 17.95%. The Akcelik model calibrated for Freeways produced a MAPE of 8.68%. In conclusion, we recommend the use of the VDF’s developed in this research for arterials and freeways for the Western Region of Puerto Rico.