As transit agencies test out emerging technologies for providing service, they encounter a challenge of having limited data compared to conventional transit operations with well-designed survey processes and an existing rider base. In such circumstances, the selection of new routes need to play a double role, not just to provide access to travelers, but also to act as “sensors” to learn how the demand responds. We developed a framework for incrementally growing out such a transit network that uses these optimal learning techniques to decide where next to expand. His tests identify the circumstances when “knowledge gradient” policies with correlated beliefs can best be fashioned into such sequential transit network design processes — a next step toward AI-driven transit network evolution that can help expedite the adoption of new transit technologies and policies.

Paper can be found here: https://www.sciencedirect.com/science/article/pii/S1366554524002989

We developed this model framework to allow policymakers and mobility market regulators to evaluate different designs and scenarios: different mixes of mobility operators and their coverage areas, pricing policies, service capacities, subsidy provision, etc., considering both fixed route and on-demand service providers. As a generalized tool, it can be used to examine special cases: mobility hub design, electric MaaS markets, first/last mile design for transit providers, cooperation and competition among mobility providers, among others. The code to the model solution algorithm can be found on Github (https://github.com/BUILTNYU/MaaS-Platform-Equilibrium).

Paper can be found here: https://www.sciencedirect.com/science/article/abs/pii/S019126152400184X

Public data for transportation planning often undersamples underserved population groups. For example, the margin of error (MOE) for population counts at census tract level for people with disabilities can make up 36% of the estimate. If we are to move towards a more equity-aware profession, the data we use need to account for this. In this work with Bingqing Liu and Farnoosh Namdarpour, we develop an algorithm to aggregate zones together to ensure sufficient data reliability for different underserved groups. We apply this algorithm for NYC to produce a set of 574 zones that result in lower MOEs compared to similarly aggregated zonal structures. We also show that this reduction in error translates to reduced variance in downstream transportation applications, such as population synthesis. The data and algorithm are publicly available and can work well with data standards like MDS.

What are the truck patterns in NYC? In this C2SMARTER Center project we (Haggai Davis, III, Hector Landes, Farnoosh Namdarpour, Hai Yang, Kaan Ozbay) synthesized truck tours from public data to provide a publicly available data set for researchers and policymakers for estimating truck VMT, evaluate impacts of truck routes, and relating those to different freight industries. It makes use of route data shared by NYC DOT (thanks Diniece Mendes, EIT A.M ASCE and team!) with borough-level screenline errors of ~10%. We can use the data to evaluate scenarios like “if truck sizes were reduced by 20%”, a preliminary assignment suggests a 49% reduction in ESALs (and impact on pavement) while increasing GHG emissions by 25%. The synthetic data are designed to be compatible with MATSim-NYC v2.0, and should be useful for evaluating off-hour deliveries, truck electrification scenarios, and multimodal last mile deliveries. The output synthetic truck tour data can be found here: https://zenodo.org/records/8000176.

Paper here: https://www.sciencedirect.com/science/article/abs/pii/S0965856424001551

One of the challenges in using a complex multiagent simulation like MATSim is that new mobility technologies and algorithms need to be custom developed as extensions in the MATSim environment. We propose using co-simulation of MATSim with other simulators and illustrate how that can allow us to, for example, evaluate two competing on-demand mobility providers in the same service region in Manhattan and their impacts on travelers’ mode and route choices under dynamic congestion. Should be of interest to MATSim users (particularly researchers that have already developed codes for the supply side simulations of their system) and agencies interested in ensembles of co-simulations for evaluating complex multi-jurisdictional policies and technologies. Contributors: Hai Yang, Haggai Davis, III, Ethan Wong, and thanks to Farnoosh Namdarpour for sharing the on-demand simulation. Funded by C2SMARTER Center, and Ethan was funded by NYU Undergraduate Summer Research Program.

Paper: https://www.sciencedirect.com/science/article/abs/pii/S1569190X24000716

In this research funded by MOIA, we (Farnoosh Namdarpour, Bingqing Liu, Nico Kühnel, Felix Zwick) investigated the prospect of adding passenger transfers to ridepooling services. The opportunity cost of committing two vehicles to a later transfer can be quite high; we compensated for this by introducing a new cost function term that can be calibrated to the operations. Results (tested on Sioux Falls and Hamburg) indicate varying levels of success that might be risky to implement at scale for general settings, but warrant further study to identify periods or hubs that may be more conducive to transfers. Should be of interest to ridepooling companies and microtransit services. Thanks to MOIA for the research support.

Srushti Rath’s last dissertation chapter on incorporating deep learning into a real options-based stochastic optimization policy for service region design: it provides a tool for decision-makers to account for uncertainty in expanding their mobility services incrementally over time. Also applicable to broader network design problems. This work was completed under Dr. Chow’s NSF CAREER project. 

Paper: https://www.sciencedirect.com/science/article/pii/S0305054823003222

With the surge in parcel deliveries especially post COVID, there is a greater need than ever to understand the trade-offs in for different delivery mechanisms. PhD student Hai (Marshall) Yang and former MS student Hector Landes (now at AECOM) developed a large scale residential parcel delivery model for NYC that approximates the vehicle miles traveled, as part of a C2SMART Center project. As NYC DOT tests out microhubs and cargo bikes and considers electric trucks, this might be able to help evaluate different scenarios. The paper was presented at the 12th International Conference on City Logistics at Bordeaux and won the Best Student Paper award. The associated data are publicly available on Zenodo: https://zenodo.org/record/7927126.

The paper can be found here: 

https://www.sciencedirect.com/science/article/pii/S2046043023000692

Building on our earlier work, we further expand the classic static dial a ride problem to model platooning between demand-responsive modular vehicles in serving customers. We invented an efficient heuristic and showed using a network of Anaheim CA how platooning can help reduce vehicle travel cost and passenger service time. Statistical tests with computational data suggest such operations and technology benefiting most from serving “enclaves” of passengers. This work should be of interest to microtransit/ridepooling or delivery operators considering more flexible vehicular technologies.

The research was funded by NSF CMMI-2022967 and USDOT #69A3551747124 via the C2SMART Center. It was also presented at the 15th International Conference on Advanced Systems in Public Transport at Tel Aviv in 2022.

Paper link: https://www.sciencedirect.com/science/article/abs/pii/S0968090X23001808