Traffic Pattern Analysis for Commuters

Wang, Haotian, Moylan, E. and Levinson, D. (2023) Ensemble Methods for Route Choice . Transportation Research part C. Volume 167, October 2024, 104803 https://lnkd.in/gZ7Q_XSA Highlights • We demonstrate the ‘best’ route choice ensemble varies with the training and testing data. • We show improvement in model performance by applying ensemble techniques to individual models. • We evaluate a set of ensemble techniques and provide recommendations for implementation and future research. • We validate the network-wide performance of the ensemble approach and individual models to predict travellers’ morning auto-based route choice by comparing the aggregate of predicted choices to observed values from loop detector records. • We recommend the use of an ensemble model employing soft voting techniques. Understanding travellers’ route preferences allows for the calculation of traffic flow on network segments and helps in assessing facility requirements, costs, and the impact of network modifications. Most research employs logit-based choice methods to model the route choices of individuals, but machine learning models are gaining increasing interest. However, all of these methods typically rely on a single ‘best’ model for predictions, which may be sensitive to measurement errors in the training data. Moreover, predictions from discarded models might still provide insights into route choices. The ensemble approach combines outcomes from multiple models using various pattern recognition methods, assumptions, and/or data sets to deliver improved predictions. When configured correctly, ensemble models offer greater prediction accuracy and account for uncertainties. To examine the advantages of ensemble techniques, a data set from the I-35 W Bridge Collapse study in 2008, and another from the 2011 Travel Behavior Inventory (TBI), both in Minneapolis–St. Paul (The Twin Cities) are used to train a set of route choice models and combine them with ensemble techniques. The analysis considered travellers’ socio-demographics and trip attributes. The trained models are applied to two datasets, the Longitudinal Employer-Household Dynamics (LEHD) commute trips and TBI morning peak trips, for validation. Predictions are also compared with the loop detector records on freeway links. Traditional Multinomial Logit and Path-Size Logit models, along with machine learning methods such as Decision Tree, Random Forest, Extra Tree, AdaBoost, Support Vector Machine, and Neural Network, serve as the foundation for this study. Ensemble rules are tested in both case studies, including hard voting, soft voting, ranked choice voting, and stacking. Based on the results, heterogeneous ensembles using soft voting outperform the base models and other ensemble rules on testing sets.

🚆 How Does a New Rail Service Impact Bus Commuters? Insights from İzmir, Türkiye 🚌➡️🚇 Fırat Enver Kesmez and Volkan Emre Uz explores how the planned metro extension in İzmir will shift travel behaviour among current bus passengers, using smart card data to track real travel patterns. 📊 Key Findings: ✅ 55% of bus passengers are expected to switch to the new metro service due to significant travel time savings—an average of 6 minutes per trip. ✅ 20% of passengers find the metro extension impractical, emphasizing the need to maintain some bus routes. ✅ 25% of bus users are "potential shifters," meaning their adoption of the metro depends on improved transfer conditions, schedules, and infrastructure. ✅ The study also highlights the role of inter-route relationships—some bus lines may need restructuring or integration to optimize public transit efficiency. By leveraging smart card data and travel time analytics, this research provides actionable insights for policymakers and urban planners to enhance the effectiveness of new rail investments and improve multimodal transit integration. 🔗 Read the full study: https://lnkd.in/g6TNg6EJ #UrbanMobility #PublicTransport #SmartData #RailInfrastructure #SustainableCities #TransportationPlanning #TravelBehaviour #BehaviourChange #ModeChoice #ModalShift

#𝐋𝐎𝐆_𝐍𝐎_𝟏𝟒𝟒 🚦 𝐓𝐫𝐚𝐟𝐟𝐢𝐜 𝐕𝐨𝐥𝐮𝐦𝐞 𝐂𝐨𝐮𝐧𝐭𝐬 𝐒𝐭𝐮𝐝𝐲 – 𝐌𝐚𝐧𝐮𝐚𝐥 & 𝐀𝐮𝐭𝐨𝐦𝐚𝐭𝐢𝐜 𝐌𝐞𝐭𝐡𝐨𝐝𝐬 𝐢𝐧 𝐏𝐫𝐚𝐜𝐭𝐢𝐜𝐞 Traffic Volume Counts study focused on collecting, analyzing, and interpreting roadway traffic patterns. This study is fundamental for traffic engineering, transportation planning, roadway safety evaluation, and future development forecasting. 📘 𝐖𝐡𝐚𝐭 𝐖𝐚𝐬 𝐃𝐨𝐧𝐞: The study focused on both manual and automatic traffic counting techniques, based on industry standards and guidance from FHWA, MUTCD, and best practices in traffic engineering. 📐 𝐓𝐞𝐜𝐡𝐧𝐢𝐜𝐚𝐥 𝐖𝐨𝐫𝐤 𝐂𝐨𝐦𝐩𝐥𝐞𝐭𝐞𝐝: 🛠️ 𝐌𝐚𝐧𝐮𝐚𝐥 𝐓𝐫𝐚𝐟𝐟𝐢𝐜 𝐂𝐨𝐮𝐧𝐭𝐬 ● Methods Used: Tally Sheets, Mechanical & Electronic Counting Boards 𝐊𝐞𝐲 𝐅𝐞𝐚𝐭𝐮𝐫𝐞𝐬: ● Conducted in 15-minute intervals during peak hours (e.g., 7:00 a.m. – 9:00 a.m.) ● Data collected included turning movements, vehicle classification, and pedestrian flow ● Observer safety ensured through appropriate PPE (hard hats, vests) and vantage point selection 🧠 𝐂𝐨𝐮𝐧𝐭 𝐏𝐥𝐚𝐧𝐧𝐢𝐧𝐠 & 𝐄𝐱𝐞𝐜𝐮𝐭𝐢𝐨𝐧 ● Chose representative timeframes to avoid anomalies (e.g., weekends, holidays, adverse weather) Followed systematic preparation including: ● Location selection based on impact studies ● Coordination with local agencies/schools/residents ● Assembled equipment checklists, recording materials, and team briefings 📊 𝐃𝐚𝐭𝐚 𝐂𝐚𝐭𝐞𝐠𝐨𝐫𝐢𝐳𝐚𝐭𝐢𝐨𝐧 & 𝐓𝐫𝐞𝐧𝐝𝐬 ● Classified vehicles as passenger cars, heavy trucks, buses, school buses Captured pedestrian counts by age group and direction ● Evaluated peak traffic volumes for use in signal timing and intersection control 🛰️ 𝐀𝐮𝐭𝐨𝐦𝐚𝐭𝐢𝐜 𝐂𝐨𝐮𝐧𝐭 𝐌𝐞𝐭𝐡𝐨𝐝𝐬 𝐀𝐩𝐩𝐥𝐢𝐞𝐝: ● Pneumatic Road Tube Systems for 24-hour data collection ● Deployed with care to avoid intersection influence zones Post-processing included: ● Data calibration & retrieval ● Error checks and summary generation ● Conversion to Average Daily Traffic (ADT) and Annual Average Daily Traffic (AADT) 🧩 𝐀𝐩𝐩𝐥𝐢𝐜𝐚𝐭𝐢𝐨𝐧𝐬 𝐨𝐟 𝐭𝐡𝐞 𝐒𝐭𝐮𝐝𝐲: ● Intersection design & control evaluation ● Traffic signal warrant analysis (as per MUTCD) ● Roadway capacity studies ● Impact analysis of proposed land use changes (e.g., new apartment developments) ● Baseline data for long-term traffic planning and zoning decisions This project deepened practical understanding of how data-driven traffic studies support smarter, safer, and more sustainable transportation systems. 📎 𝐑𝐞𝐟𝐞𝐫𝐞𝐧𝐜𝐞 𝐒𝐭𝐚𝐧𝐝𝐚𝐫𝐝𝐬: ● ACI Manual on Transportation Engineering Studies ● MUTCD (FHWA) ● Local & regional DOT guidelines #𝐓𝐫𝐚𝐟𝐟𝐢𝐜𝐄𝐧𝐠𝐢𝐧𝐞𝐞𝐫𝐢𝐧𝐠 #𝐔𝐫𝐛𝐚𝐧𝐏𝐥𝐚𝐧𝐧𝐢𝐧𝐠 #𝐓𝐫𝐚𝐧𝐬𝐩𝐨𝐫𝐭𝐚𝐭𝐢𝐨𝐧𝐏𝐥𝐚𝐧𝐧𝐢𝐧𝐠 #𝐓𝐫𝐚𝐟𝐟𝐢𝐜𝐕𝐨𝐥𝐮𝐦𝐞𝐂𝐨𝐮𝐧𝐭 #𝐌𝐚𝐧𝐮𝐚𝐥𝐂𝐨𝐮𝐧𝐭𝐬 #𝐀𝐃𝐓 #𝐀𝐀𝐃𝐓 #𝐌𝐨𝐛𝐢𝐥𝐢𝐭𝐲𝐒𝐨𝐥𝐮𝐭𝐢𝐨𝐧𝐬 #𝐅𝐢𝐞𝐥𝐝𝐃𝐚𝐭𝐚𝐂𝐨𝐥𝐥𝐞𝐜𝐭𝐢𝐨𝐧 #𝐈𝐧𝐟𝐫𝐚𝐬𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞𝐏𝐥𝐚𝐧𝐧𝐢𝐧𝐠

Traffic Patterns, Human Behavior & The Tale of Two Roads In India, a new expressway is rarely just a road. It’s a fresh choice, a new promise — and a live experiment in how people make decisions behind the wheel. When a competing route opens alongside an existing highway, traffic doesn’t simply “shift.” It behaves like water — flowing to the path of least resistance, yet shaped by habits, perceptions, and sometimes pure emotion. Why Drivers Choose the Routes They Do: A. Predictability over speed – Many prefer a slightly longer but consistent route over one with unpredictable delays. B. Cost sensitivity – Toll rates, fuel efficiency, and trip length impact fleets and personal vehicles differently. C. Comfort & safety – Smooth surfaces, wide lanes, and better lighting often tip the scales. D. Habit inertia – Old routes keep loyalty long after new options emerge. Impact on Existing Highways: When a new expressway launches: A. Premium, time-sensitive traffic often shifts quickly. B. Freight may stick to old routes if tolls are high. C. Local and short-haul users keep older corridors alive. Case Study: Yamuna Expressway vs NH-19 (Delhi–Agra) Tourists & high-end buses moved early to the expressway for speed and comfort. Freight operators stayed on NH-19 for cost and convenience. Roadside businesses adapted to focus on local customers. Today: The expressway handles most premium passenger traffic; NH-19 still carries a significant amount of freight. Key Insight: A new road doesn’t erase the old — it redefines it. How to Predict Traffic Shifts Right: A. Origin–Destination mapping – Know who benefits most from the new route. B. Behavioral modelling – Factor in toll elasticity, perceived safety, and driver habits. C. Early usage trials – Temporary toll changes or soft launches reveal real patterns. D India-specific context – Festivals, freight bans, harvest seasons, and roadside traditions matter. Why This Matters Now: With expressways like Delhi–Jaipur, Delhi–Mumbai, Mumbai–Nagpur, and others coming online, accurately predicting adoption is just as critical to their success as building the infrastructure itself. Pricing will shape uptake. Services (rest stops, EV charging) will win loyalty. Integrating old and new corridors will protect local economies. Traffic Adoption Curve: Illustration showing how usage evolves: A. Month 0–3: Curiosity spike. B. Month 4–9: Price sensitivity and habit play tug-of-war. C. Month 10–18: Stable split between expressway and old highway. Bottom line: Roads shape people, but people shape roads. Predicting it right means blending engineering with human psychology. Disclaimer: These are my personal observations and interpretations of traffic behavior. They do not reflect the views of the organization, and are not based on official traffic studies. Actual patterns may vary depending on local conditions and traveler choices.