Big Data: Future of Transportation Studies

About a year ago, Sain published a blog discussing the use of big data as an alternative for traffic data collection during COVID-19. At that time, travel behavior changed, the number of vehicles on the roads had decreased, and peak hours were not the same when compared to pre-pandemic times.

Since manually collecting reliable traffic counts was no longer an option, traffic engineers and transportation planners found big data to be an excellent alternative for data collection for transportation projects requiring traffic counts, such as proposed development projects, signal timing adjustments, public roadway improvements, and traffic forecasting. Big data allowed traffic counts to be obtained retroactively, making requests for data from pre-pandemic changed travel patterns a possibility.

As life returns to normal conditions and restrictions are lifted, big data appears to be a good source for data collection moving forward, especially in the transportation planning and safety fields. Several metrics can be obtained with big data: Annual Average Daily Traffic (AADT), trip origin-destination, trip length, trip route, trip purpose, speed, travel time, turning movement counts (TMCs), and vehicle hard braking data are a few examples. Several applications of big data to transportation studies are presented below:

  • Safety studies: traditional safety analysis has been focused on a “reactive” approach. Hotspots are identified based on historical crash data, which requires at least three to five years of crash data. Locations with a high number of crashes, or severe crashes, are candidates for safety improvements. After improvements are implemented, another three to five years of crash data are required to evaluate the safety effectiveness of the countermeasures. Big data is an excellent alternative to a “proactive” safety analysis, which means high-risk locations can be identified even before crashes occur. Hard braking data, for example, can be obtained from big data, allowing safety improvements to be made at high-risk locations. There are several examples of studies evaluating the benefits of big data to safety analysis.  
  • Transportation forecasting: estimating future traffic volume begins with using reliable existing traffic data representative of typical conditions. The typical manual data collection of TMCs only includes observations on a specific day for a few hours in the morning and evening. The collected data is then assumed to represent the entire year, which is not always true. One of the major benefits of big data is the ability to look at information over the whole year. Further, historical counts are rarely available for off-peak times on weekdays or during the weekends. Big data allows data aggregation by any hour of a day and any day of the week.
  • Before-after studies: big data can be obtained retroactively as opposed to scheduling future data collection. As an example, a certain jurisdiction conducts a corridor study with roadway segment and intersection improvements. Big data can summarize what travel time and speeds were before corridor improvements were implemented and compare those to what travel times and speeds are after corridor improvements are in place.

As we have navigated the uncertain conditions of the COVID-19 pandemic, we have learned so much. Being able to work together to find solutions to overcome adversities was probably the main lesson learned. People worked hard to adapt to fluid conditions, and it was no different in the transportation field. Big data allowed transportation professionals to keep important projects going during the pandemic, and it now emerges as a promising alternative to improve transportation studies moving forward.