NYC Ferry has connected underserved waterfront neighborhoods to the broader transit network since May 2017, running seven routes across the five boroughs. On-time performance varies considerably by route, time of day, and environmental conditions — but the operations team has little visibility into likely delays before a trip begins.
Working with the NYC Economic Development Corporation (NYCEDC), our team of four graduate students from the Master of Urban Spatial Analytics and Master of City Planning programs produced three deliverables. We built a historical analysis identifying the operational, environmental, and demand-side factors most associated with delays, a 48-hour predictive model that forecasts stop-level delay risk using ridership patterns, vessel traffic, and environmental conditions, and an interactive web application that operationalizes those predictions for real-world use. The goal is to give operators and riders a clearer picture of where and when delays are likely to happen, far enough in advance to act on it.
