Studying the trends in traffic collisions during Covid-19

Kokoutse Xavier Doh and Dr. Daehan Kwak Kean University, School of Computer Science and Technology, Union, NJ 07083

Recent studies at NHTSA show that in average 6 million of car accidents are reported among which, 6% are fatalities, 27% are non-fatalities and 72% are damaged properties. More than 38,000 people die every year and at least 4.4 million are injured. The typical causes of crashes that result in death are: 40% alcohol, 30% speeding and 33% reckless driving. In fact, federal studies found that the economic and societal harm from traffic accidents cost about $871 billion each year. Covid-19 has not only affected the United States economically, but also, in many other ways. Several workers are urged to work remotely, states have ordered lockdown and social distancing to mitigate the spread of the virus. A case study is performed with emphasis on New York City, collecting vehicle collisions data from 6 months prior March 2020 up to the ones that recently occurred. Based on these information, the initial data provided by the NYPD (New York Police Department) and obtained from the NY open source data website, is engineered, trimmed, and exported as a CSV file, which is used for this study. To convert the raw data into an understandable information format for humans to easily read and deduct a conclusion from, the data is imported to a data visualization platform called Tableau. Several charts are implemented to visualize metrics such as: the total number of accidents occurred before and after the NYC lockdown period, rate of collisions over the months, injuries versus deaths, location concentration percentage of collisions and factors related to these collisions. The importance of the data visualized on traffic regarding pre-and post-shutdown will shed meaningful information, that can support private and public sectors in their decision making on how to manage their business or resources.

Additional Abstract Information

Presenter: Kokoutse Doh

Institution: Kean University

Type: Poster

Subject: Computer Science

Status: Approved

Time and Location

Session: Poster 5
Date/Time: Tue 12:30pm-1:30pm
Session Number: 4029