Research Projects
Current Projects
-
Incorporating New Mobility Options into Travel Demand Forecasting and Modeling, NCHRP 20-102(29), Funded by the Transportation Research Board of the National Academies. Role: Co-Principal Investigator.
-
Evaluating the Impacts of Real-Time Warnings and Variable Speed Limits on Safety and Travel Reliability during Weather Events, NCHRP 03-142, Funded by the Transportation Research Board of the National Academies. Role: Senior Personnel and Project Leader. Collaborative project.
-
Predicting Real-time Population Behavior during Hurricanes Synthesizing Data from Transportation Systems and social media, Funded by National Science Foundation (NSF). Role: Senior Personnel and Project Leader; Collaborative project.
-
The Effect of Vehicle Mix on Crash Frequency and Crash Severity, NCHRP 22-49, Funded by the Transportation Research Board of the National Academies. Role: Senior Personnel and Project Leader.
Completed Projects
Enhancing Non-Motorized Safety by Simulating Non-Motorized Exposure using a Transportation Planning Approach
Sponsor: SAFER-SIM UTC
Develop a transportation planning simulation framework to generate non-motorists’ exposure information for crash prediction models. The evaluated exposure measures are incorporated in examining non-motorists’ safety to devise more evidence based policy implications for improving overall safety and activities related to non-motorized modes of travel.
The Effect of Vehicle Mix on Crash Frequency and Crash Severity
Sponsor: NCHRP
Develop and validate a statistically valid predictive methodology to quantify the effect of vehicle mix on crash frequency and severity for various facility types. Further, we developed a spreadsheet tool for practitioners to quantify the effect of vehicle mix on safety performance across the range of highway activities including planning, design, operations, and safety management.
Development and Application of Crash Severity Models for Highway Safety
Sponsor: NCHRP
We identified gaps and opportunities within the current HSM procedures. Then, we explored newer alternative approaches in developing crash prediction models considering different crash severity types and access their performances to identify the best approach. Finally, develop guidance document with protocols for the use and application of the proposed severity models.
RAD
Disaggregate Realistic Artificial Data (Rad) Generator - Design, Development and Application For Crash Safety Analysis
Sponsor: FHWA
Build a high resolution disaggregate data generation process that mimics crash occurrence on transportation facilities. The framework operated at the trip level and incorporates influence of full range of crash contributing factors. The ARD framework is general enough to be able generate crashes for all roadway facility types including segments and junctions.
Evaluating the benefits of multi-modal investments on promoting travel mobility in Central Florida
Sponsor: FDOT
Evaluate the benefits of multi-modal investments on promoting travel mobility in Central Florida region. Developed a micro-simulated trip-based framework to accommodate the potential adoption of non-auto modes in response to the growing emphasis of non-auto mobility – public transit, pedestrian, and bicyclist modes in Central Florida’s urban regions.
Integrated Freeway/Arterial Active Traffic Management
Sponsor: FDOT
Explored the road users’ preferences of private vehicle drivers in and around greater Orlando region through a web-based state preference survey. Based on the responses, we developed a MNL model to understand the effect of different attributes on route choice behavior. The survey results provide route diversion information as part of the decision support system being developed.
Improving Vulnerable Road Users’ Safety Based on Computer-Vision Technologies and High-fidelity Hybrid Traffic Forecasting Tool for Urban Transportation Networks using Emerging Datasets
Sponsor: FDOT
Develop a hybrid travel demand and data-driven traffic forecasting tool that takes intersection-level historical traffic movement data and travel demand observations at various spatio-temporal resolutions and employs host of variables as inputs along to predict future traffic movements. Further, utilize the benefits of convolutional neural network in capturing cross-correlation among different spatial features of a transpiration network
Evaluating Community Building Effectiveness of Transportation Investments: Using Traditional and Big Data Oriented Analytical Approaches
Sponsor: FDOT
Develop and implement a framework to compare the changes in measure of effectiveness (MOEs) across scenarios to identify benefits to the Central Florida region. Additionally, we conduct an extensive knowledge transfer activity through webinars and supporting manuals: provide step-by-step guidance on the various data preparation, data download and data analysis tasks conducted for the project .
A pragmatic multi-objective planning approach for medium and long range projects
Sponsor: FDOT
Summarize an exhaustive review of transportation and urban planning studies that conducted project planning forecasts. Based on that, we compiled the data sources encompassing various dimensions that are useful for conducting periodic project evaluation. Data sources are identified for three major projects: I-4 Ultimate project, SunRail and Wekiva Parkway.
Quantification of health risks and benefits of future transportation plans, Collaborative Health Research Project
Sponsor: CIHR-NIH
The project aims at developing a transportation model for Greater Montreal (Canada) to predict the traffic and trips by various transportation modes for future years. The final objective is comparing the number of trips, distances by mode, air pollution and health impacts of Business as Usual (BAU) with varying scenarios.