Key points are not available for this paper at this time.
The purpose of this project is to predict the number of taxi trip that goes unmet in NYC by using available trip data generated by Taxi Cabs. The expected outcome from this project is to identify the areas in NYC that experiences unmet demand and to identify socio-economic features that impact trip demand. The project was broken down into 2 phases. To account for daily trip trends the study is broken down in 7 periods on Monday to Saturday and 4 periods on weekends. In the rst phase 3 approaches were developed to identify census tract where the demand of taxi is not met. The rst approach was to compare the monthly total pickup and drop-of counts. The second approach identied the underserved census tracts by nding those with the least number of vacant taxis within a given time duration. In the third approach, the rate of change Uber pickup was compared with the rate of change of combined pick up of all taxi services over a 6 months period. The nal output was: 358 unmet trip demand CT that had pick-ups to drop-offs ratio lower than 25%, Less than 2 vacant taxi a minute and a Uber growth rate more than 4%. These areas were in North-east Queens, South Brooklyn and North of the Bronx. In Phase II, the socio-economic features were ranked to understand their inuence on taxi demand. Out of the 11 selected features the top 5 features were neighborhoods with high number of offices and employees, residential areas, income and rent. Using the 11 features as predictors and pick-up from CT with satisfactory as target variable a model was developed to predict demand for underserved areas. The model predicted overall 92,000 trips per day goes unmet in NYC which is 4.2% of the total current daily trips.
Ahmed et al. (Fri,) studied this question.