Details:

Creators:
Kaseko, Mohamed ; Teng, Hualiang (Harry) ; Kumaresan, Vidhya ; Chigurupati, Saketh
Kaseko, Mohamed ; Teng, Hualiang (Harry) ; Kumaresan, Vidhya ; Chigurupati, Saketh Less -
Corporate Creators:
University of Nevada Las Vegas. Department of Civil and Environmental Engineering and Construction
Subject/TRT Terms:
[+]
Accident Analysis Dashboard Data Analysis Fuel Consumption Generalized Linear Models GLMs Regression Analysis Traffic Congestion Travel Time
Publication/ Report Number:
001-13-803
Resource Type:
Tech Report
Geographical Coverage:
Nevada
Corporate Publisher:
Nevada. Dept. of Transportation
NTL Classification:
NTL-HIGHWAY/ROAD TRANSPORTATION-HIGHWAY/ROAD TRANSPORTATION;NTL-OPERATIONS AND TRAFFIC CONTROLS-Congestion;
Abstract:
Incidents are a major source of non-recurring congestion on freeways. In addition to costing millions of dollars in the loss of life, injuries and property damage, traffic incidents also cause additional losses due to the resulting traffic congestion, delay and energy consumption. Depending on the severity of an incident, in terms of the number and location of travel lanes blocked and the duration of the incident, the resulting congestion can cause significant additional traffic delays, travel time, and associated additional fuel consumption and vehicle emissions. The objective of the proposed research is to model and quantify the impacts of freeway incidents on measures of effectiveness including system-wide traffic travel times, fuel consumption and vehicle emissions. Statistical regression models are calibrated that relate freeway travel times, fuel consumption and emissions as functions of incident characteristics that include incident duration, number of lanes blocked and corresponding non-incident traffic characteristics. One year worth of data for a section of the northbound I-15 freeway in Las Vegas metropolitan area is used for the study. The data is retrieved from Dashboard, an interactive website maintained by RTC’s FAST. Non-linear regression models are calibrated for each impact variable using the statistics software package R. Models are calibrated for (i) excess travel time per vehicle (ii) excess vehicle-hours of travel (iii) excess fuel consumption and (iv) excess vehicle emissions (CO2, CO, NOx and PM10) for all vehicles over the spatial and temporal extent of incidents. The full set of predictor variables used included incident duration, number of lanes blocked, lane-minutes of blockage (product of incident duration and number of travel lanes blocked), location of blocked lanes, ratio of lanes blocked, peak/off-peak period, day-of-week (weekday versus weekend), traffic volume, speed and density for non-incident conditions over the corresponding spatial and temporal extents of incidents. The statistical model results indicate, as expected, that the most significant predictor variables are the incident duration, number of lanes blocked and the non-incident traffic density. In certain models, the incident duration and lanes blocked were replaced by the product of the two, namely, the lane-minutes of blockage. The resulting statistical functional forms are the Gaussian Single-Log and Double-log Generalized Linear Models (GLMs). Use of the models is demonstrated by showing examples of using the equations to compute the impact of an average incident. The results show, for example, that an average incident that has one travel lane blocked on the section of the freeway modeled results in approximately 149.2 excess vehicle-hours of traveland 41.45 gallon of excess fuel consumed by impacted vehicles. Further analysis using elasticity derived equations can be done to estimate marginal impacts with respect to small changes in the values of the predictor variables, such as the incident duration and n umber of travel lanes blocked. Such analysis can be used for planning purposes and for evaluation of the overall performance of a freeway network, as well as for benefit-cost evaluation of incident management projects.
Format:
PDF
Collection(s):
US Transportation Collection
Main Document Checksum:
[+]
urn:sha256:70046375fe45ada6f1ac314bfd7f6230787383ee840b847e6146814042186f2d
Download URL:
https://rosap.ntl.bts.gov/view/dot/34880/dot_34880_DS1.pdf
File Type:

Supporting Files

• No Additional Files

More +