Analytical Model for Traffic Congestion and Accident Analysis

Traffic congestion and road accidents have been important public challenges that impose a big burden on society. According to the statistics report from the Association for Safe International Travel (ASIRT), an average American lost $1,377 and 99 hours sitting in traffic in 2019, over 37,000 people die in road crashes each year, and 2.35 million are injured or disabled that costs $230.6 billion per year. The analysis of traffic congestion and road accidents is very complex as they not only affect each other, but they are also affected by many other factors. In this research, we use the US Accidents data from Kaggle that consists of 3 million records of car accidents from February 2016 to December 2019, and 49 variables for the study. We propose to use statistical techniques and machine learning algorithms including Logistic Regression, KNN, Decision Tree Classifier, and Random Forest Classifier to process and train a large amount of data to obtain predictive models for traffic congestion and road accidents. The proposed predictive models will incorporate multiple environmental parameters and are expected to be more accurate with the input of real-time information. The proposed models will assist people in making smart real-time transportation decisions to improve mobility and reduce accidents.