In the vibrant landscape of New York City, the yellow taxi industry stands as an iconic symbol and a lifeline for countless city residents and tourists. The availability and effective distribution of yellow taxis throughout the day play a pivotal role in meeting the ever-evolving demands of urban transportation.
In our project, we focus on analyzing the time series demand of New York yellow taxis on an hourly and monthly basis. We will experiment with classical decomposition, smoothing techniques, and (S)ARIMA models to select the most robust forecasting model to accurately predict ride demand. Our aim is to uncover the underlying trends, seasonality, and patterns, especially during peak hours, to gain valuable insights into the city's transportation dynamics. By doing so, we strive to make recommendations to not only enhance the operational efficiency of taxi services but also contribute to a more efficient and responsive urban transportation system in New York City. Taxi demand forecasting offers real-time applications with significant benefits and impacts on the transportation ecosystem. It enables taxi service providers to allocate resources efficiently, implement dynamic pricing, and optimize driver schedules. It aids traffic management, reduces empty cruising, and promotes environmental sustainability. For passengers, it reduces waiting times and improves convenience.
Our project focuses on several key questions related to taxi service and passenger behavior. We aim to determine the peak daily demand for taxis and whether it varies day-to-day. We will identify the busiest month and track demand trends within it. We'll pinpoint the most revenue-generating ride types and predict when longer rides are booked. Lastly, we'll explore common destinations for our taxi service users. These questions are the basis for our research and analysis. The dataset originates from the official website of the City of New York and has been provided by authorized technology providers participating in the Taxicab & Livery Passenger Enhancement Programs for the NYC Taxi and Limousine Commission (TLC). Our primary variables include total revenue, time, number of rides, distance, location, passenger count etc. KPIs, we look to address include Number of rides in an hour/day, Revenue/ride, Revenue/min, and Revenue/mile.
The NYC taxi time series analysis offers vital insights for strategic decision-making. This includes identifying peak hours for potential surge pricing, which informs resource allocation and pricing strategies. Demand forecasting ensures efficient driver and vehicle deployment, upholding service quality. Additionally, operational efficiency can be improved by identifying and addressing inefficiencies through responsive measures.