Lead Presenter Biography

Mohammad is a PhD student at Arizona State University, specializing in the field of transportation engineering. His research focuses on large-scale traffic analysis, multi-resolution modeling, and energy estimation by utilizing data-driven techniques and advanced computational tools.

Co-Authors

Presentation Description

The significant increase in energy consumption and greenhouse gas emissions (GHG) causes transportation agencies to seek new active traffic management systems. Many organizations consider integrated corridor management strategies to mitigate congestion and improve air quality, while others encourage public transit and non-motorized transportation. In this regard, eco-routing or green routing in route guidance provision is receiving increasing attention from the field of green transportation. The idea of green routing is to help drivers make greener choices about their routes by providing the most eco-friendly route in terms of minimum emissions.
Microscopic traffic simulation tools have been frequently used to estimate vehicle emissions by analyzing the driving speed and acceleration characteristics or profiles on per vehicle and second-by-second basis. Although a high-fidelity traffic simulator is ideal for studying individual vehicle delays, microscopic simulation can be computationally intensive and frequently requires a wide range of geometric data and driving behavior parameters, which can be challenging to calibrate, especially to produce high-fidelity emissions estimates. Therefore, selecting an appropriate temporal and spatial resolution, traffic assignment, and emission estimation models is of utmost significance to establish a trade-off between computational efficiency and estimation accuracy.
This research addresses the eco-system optimal dynamic traffic assignment (ESODTA) problem, which aims to find system optimal eco-routing or green routing flows that minimize total vehicular emission in a congested network. We present a Multi-Resolution Modeling (MRM) framework that integrates mesoscopic Dynamic Traffic Assignment (DTA) and a fluid-based queue model to capture the relationship between delay and link emissions at different spatial scales and various temporal resolutions. Using the Queue Volume to Delay Functions (QVDFs) and a simplified MOVES (Motor Vehicle Emission Simulator) model, we hope to obtain an analytical form of the average emission function in terms of a queue evolution process for a range of inflow demand over capacity ratio. Then, time-dependent demand patterns, which result in a closed-form time-dependent queue length and speed profile, can be used to expand the vehicle's emission calculation to a day-based analysis (instead of a period-based analysis). Total Vehicle Miles Travelled (VMT) and Vehicle Hours Travelled (VHT) can also be calculated precisely with an accurate time-dependent speed profile.
Similar to MOVES, this simplified MOVES model uses the VSP-to-Operating Mode conversion table and considers average emissions rates stratified by vehicle type, age, and vehicle operating mode. VSP is a function of vehicle speed, road grade, and acceleration, accounting for kinetic energy, rolling resistance, aerodynamic drag, and gravity. Unlike MOVES, which models more than ten vehicles types, this simplified model, known as MOVES Lite, only considers a limited number of vehicle types that represent 95% of the on-road fleet, namely passenger cars, passenger trucks, light commercial trucks, single unit short-haul trucks, and combination long-haul trucks. MOVES Lite can significantly reduce the number of source bins by using a limited number of vehicle types, dramatically decreasing the complexity of the emission rate search process and improving calculation efficiency. 

Presentation File