How Sensors and Analytics are Transforming Traffic

Written By: Guest Author
  • 1/19/2017

By Judith Hurwitz, President & CEO, Hurwitz and Associates

In cities everywhere, traffic congestion leads to frustration, lost productivity, inefficiencies in commerce and delays in emergency responses. Traffic is a challenge that has frustrated planners, administrators, and citizens for decades; progress, when it has been achieved, has been very limited.

Now, however, a confluence of better and more affordable sensor technology, improved networking, and powerful real-time analytics and machine learning are putting major city improvements within reach.

Cities can be defined as layers of complex systems that have to work in collaboration with each other to function successfully. Computer models and day-to-day experience can provide some guidance for helping things to run smoothly, but only some. There has simply been too much complexity and too many variables. Even an individual vehicle – itself a complex system – may be subject to hard-to-predict failures that can upend the best planning and the most sophisticated attempts at traffic management. In short, one flat tire can cause chaos across a metropolitan region.

The key question is how do you move from the normal chaos of traffic to a controlled and managed environment? The answer lies in data and the evolution of machine learning and analytics.

In recent years, the cost of sensors that measure everything from speed to temperature, location, and acceleration has decreased dramatically. In addition, sensor battery life has increased, making longer term monitoring of more things far more practical. And the networks that support communication between sensors and between sensors and gateway devices have become ever more robust. We are therefore seeing more and more sensors added to both vehicles and to urban infrastructure. These sensors are making it possible for planners and administrators to collect and analyze data both for the long term and in near real time to support minute-by-minute decision making.

This is completely changing operational approaches.

Until now, we could only collect data as an afterthought to understand what happened in the past, we have not been able to apply data through a machine learning process to solve problems as they happen. Accidents, weather events, or a public or private activity generating unusual volumes of traffic were largely beyond the scope of anything but a tactical response such as simply dispatching extra law-enforcement units.

By ingesting and analyzing data from these complex sensor-based systems, city leaders are better able to take action as a situation unfolds. For example, emergency dispatchers can help emergency vehicles bypass choke points to get to an incident quickly. In addition, traffic signals can be actively managed and coordinated throughout the day and night to make the flow of traffic most efficient. Citizens can also get live updates on public transportation, private taxis and ride sharing options to help plan and route their trips based on their speed and budget priorities.

These capabilities are rapidly moving from innovative concept to reality. Cities such as Toronto in Canada have adapted a reinforcement learning traffic system that manages traffic. The system is able to manage traffic signals proactively when traffic congestion begins to impact the movement of vehicles in the city. The result has been a 40 percent reduction in delays in the downtown area. Similarly Rio de Janeiro in Brazil implemented a city and traffic monitoring system to tackle the challenges associated with hosting the Summer Olympics. The Rio system monitored everything from traffic to weather and even assesses the amount of standing water in streets. The goal of the system was to help city leaders make fast, data-driven decisions, minimize citizen disruptions, and efficiently deploy city resources.

The value of applying machine learning and big data analytics to traffic problems in cities is an example of how sensor data can dramatically improve the way metropolitan areas are managed – gaining an advantage from complexity rather than succumbing to it.

Systems interacting with systems through sensors and machine learning help ensure the right approaches and processes to transform chaos into order. The impact can be dramatic, helping cities transform themselves into more livable environments. 

Related Articles:

  • CAD
  • Industrial Connectivity
  • Industrial Internet of Things

About the Author

Guest Author

Product Lifecycle Report guest authors are industry thought leaders on topics and insights related to the IoT. If you are interested in becoming a guest author, please contact Michelle Hopkins at