United States: Reducing Transportation CO₂ and Greenhouse Emissions
February 2025
Project Type: Sustainable Transportation & Emissions Reduction Research
Focus Areas: CO₂ Emissions, Online Shopping Behavior, VMT Reduction, Urban Planning, Climate Policy
Publication Details for Transportation Research Paper:
Published In: Transportation Research Part D: Transport and Environment
Volume: 139, February 2025, Article 104567
Authors: Manreet Sohi, Patrick Loa, Basar Ozbilen, Xiatian Iogansen, Yongsung Lee, Giovanni Circella
DOI: 10.1016/j.trd.2024.104567 . If you do not have full access to view the paper, feel free to contact me for a copy.
Overview
Transportation is one of the largest contributors to global greenhouse gas emissions, and the COVID-19 pandemic dramatically reshaped how people move and shop. This project set out with a clear goal: to identify actionable policies that reduce CO₂ emissions by understanding how online shopping influences vehicle travel behavior. Using a LightGBM machine learning model on California household behavior, we analyzed how food and non-food e-commerce affected transportation patterns across different neighborhood types. By capturing changes in behavior during a pivotal time, this study offers data-backed solutions to help cities and governments reduce VMT and meet climate targets. As lead author, I was responsible for the full technical development of the machine learning model in Python, statistical analysis in R, SHAP-based interpretation of results and ATE simulation of the policy recommendations. I also led the writing process, from drafting to revising, and worked closely with postdoctoral researchers to refine our methodology and produce actionable, policy-relevant findings.
The ultimate goal: Turn behavioral insights into smart, scalable strategies for cutting transport-related emissions in a digital economy.
Key Findings
Machine Learning Model Performance: The LightGBM model demonstrated superior predictive accuracy compared to traditional regression methods, achieving a 15% lower Mean Absolute Error (MAE) and capturing complex non-linear relationships between online shopping behavior and VMT that linear regression failed to identify.
Online Shopping Effects: revealed contrasting impacts - Food-related online shopping (groceries, restaurant orders) showed a substitution effect. Non-food online shopping (clothing, electronics) demonstrated a complementary effect.
Built Environment Influence: Higher density neighborhoods (>10,000 people/sq mile) and frequent transit service (>20 trips/hour) were associated with 25% lower shopping VMT compared to low-density areas, highlighting the importance of urban form in sustainable transportation patterns.
Policy Recommendations
Expand public transit infrastructure and service frequency to support high-density, mixed-use development that reduces car dependency.
Develop centralized delivery hubs near transit corridors to minimize the number of last-mile delivery trips.
Prioritize walkable, mixed-use neighborhoods with easy access to retail and active transportation options like biking and walking.
Offer incentives for electric and low-emission vehicle adoption across both personal use and delivery fleets.
These strategies aim to reduce shopping-related vehicle miles traveled (VMT) and lower greenhouse gas emissions while maintaining the convenience of online shopping.
Anticipated Impact: This research offers a data-driven framework to guide urban planners, policymakers, and sustainability professionals in reducing shopping-related transportation emissions. By identifying how different types of online shopping influence travel behavior—and how built environment factors can mitigate VMT—the study provides actionable insights for designing low-carbon, people-centered cities.
The paper was presented at the Transportation Research Board (TRB) Annual Meeting 2025 in Washington, D.C., and the 3RFM Research Workshop in November 2024. These presentations enabled cross-sector dialogue and highlighted the potential for integrating machine learning into transportation planning to accelerate progress toward national and local CO₂ reduction goals.
End result: This research offers data-driven policy pathways to cut transportation-related CO₂ emissions by aligning online shopping trends with sustainable urban planning.