Data-Driven Prediction and Optimization of Energy Use for Transit Fleets of Electric and ICE Vehicles

Due to the high upfront cost of electric vehicles, many public transit agencies can afford only mixed fleets of internal combustion and electric vehicles. Optimizing the operation of such mixed fleets is challenging because it requires accurate trip-level predictions of electricity and fuel use as well as efficient algorithms for assigning vehicles to transit routes. We present a novel framework for the data-driven prediction of trip-level energy use for mixed-vehicle transit fleets and for the optimization of vehicle assignments, which we evaluate using data collected from the bus fleet of CARTA, the public transit agency of Chattanooga, TN. We first introduce a data collection, storage, and processing framework for system-level and high-frequency vehicle-level transit data, including domain-specific data cleansing methods. We train and evaluate machine learning models for energy prediction, demonstrating that deep neural networks attain the highest accuracy. Based on these predictions, we formulate the problem of minimizing energy use through assigning vehicles to fixed-route transit trips. We propose an optimal integer program as well as efficient heuristic and meta-heuristic algorithms, demonstrating the scalability and performance of these algorithms numerically using the transit network of CARTA.

Public Transit Development Predicts Spatial Distribution of Dengue Virus Incidence in Medellín, Colombia

Dengue is a growing global threat in some of the world’s most rapidly growing landscapes. Urbanization and human movement affect the spatial dynamics and magnitude of dengue outbreaks; however, precise effects of urban growth on dengue is not well understood because of a lack of sufficiently fine-scaled data. We analyzed nine years of address-level dengue case data in Medellin, Colombia during a period of public transit expansion. We correlate changes in the spread and magnitude of localized outbreaks to changes in accessibility and usage of public transit. Locations closer to and with a greater utilization of public transit had greater dengue incidence. This relationship was modulated by socioeconomic status; lower socioeconomic status locations experienced stronger effects of public transit accessibility and usage on dengue incidence. Public transit is a vital urban resource, particularly among low socioeconomic populations; these results highlight the importance of public health services concurrent with urban growth.

Facing the future of transit ridership: which riders bought a car; who is planning on riding less?

Public transit agencies face a transformed landscape of rider demand and political support as the COVID-19 pandemic continues. We explore people’s motivations for returning to or avoiding public transit a year into the pandemic. We draw on a March 2021 follow up survey of over 1,900 people who rode transit regularly prior to the COVID-19 pandemic in Toronto and Vancouver, Canada, and who took part in a prior survey on the topic in May, 2020. We model how transit demand has changed due to the pandemic, and investigate how this relates to changes in automobile ownership and its desirability. We find that pre-COVID frequent transit users between the ages of 18-29, a part of the so-called “Gen Z,” and recent immigrants are more attracted to driving due to the pandemic, with the latter group more likely to have actually purchased a vehicle. Getting COVID-19 or living with someone who did is also a strong and positive predictor of buying a car and anticipating less transit use after the pandemic. Our results suggest that COVID-19 heightened the attractiveness of auto ownership among transit riders likely to eventually purchase cars anyways (immigrants, twentysomethings), at least in the North American context. We also conclude that getting COVID-19 or living with someone who did is a significant and positive predictor of having bought a car. Future research should consider how the experiencing of having COVID-19 has transformed some travelers’ views, values, and behaviour.

Bus Travel Time Prediction: A Comparative Study of Linear and Non-Linear Machine Learning Models

Congested roads are a global problem, and increased usage of private vehicles is one of the main reasons for congestion. Public transit modes of travel are a sustainable and eco-friendly alternative for private vehicle usage, but attracting commuters towards public transit mode is a mammoth task. Commuters expect the public transit service to be reliable, and to provide a reliable service it is necessary to fine-tune the transit operations and provide well-timed necessary information to commuters. In this context, the public transit travel time is predicted in Tumakuru, a tier-2 city of Karnataka, India. As this is one of the initial studies in the city, the performance comparison of eight Machines Learning models including four linear namely, Linear Regression, Ridge Regression, Least Absolute Shrinkage and Selection Operator Regression, and Support Vector Regression; and four non-linear models namely, k-Nearest Neighbors, Regression Trees, Random Forest Regression, and Gradient Boosting Regression Trees is conducted to identify a suitable model for travel time predictions. The data logs of one month (November 2020) of the Tumakuru city service, provided by Tumakuru Smart City Limited are used for the study. The time-of-the-day (trip start time), day-of-the-week, and direction of travel are used for the prediction. Travel time for both upstream and downstream are predicted, and the results are evaluated based on the performance metrics. The results suggest that the performance of non-linear models is superior to linear models for predicting travel times, and Random Forest Regression was found to be a better model as compared to other models.