Analysis of Trend in Traffic Flow Using Computational and Statistical Approaches

Predicting traffic trends is an important subset of civil engineering. Engineers and administrators of a township must construct a comprehensive strategy to optimize traffic patterns prior to the construction of buildings and roads. This paper analyzes traffic trends to optimize road design and traffic management. The authors collected traffic data from New York City and Bronx Borough including peak travel hours, and used a distribution method to map out overall traffic patterns. The data were plotted using time-series decomposition in Python and MATLAB® programming. Traffic volume for each time period was analyzed to enhance the accuracy of traffic data. Scatter diagrams and linearity analyses were performed to analyze traffic trends. The ARIMA (Autoregressive Integrated Moving Average) model was used to model the time series nonlinear forecasting of noisy traffic data.