Big Data to Reduce Urban Traffic Accidents

Utilizing data collected from sensors installed along urban transportation routes to improve road safety.

Creating a Smarter Future

Connecting the Dots

Big Data to Reduce Urban Traffic Accidents

To reduce or, better yet, eliminate the road death toll, what is needed is a focused, effective response, and a concrete plan of action. Recently, data collected from sensors installed throughout urban roads have begun to be used to provide some kind of solution.


While improvements in vehicle safety mechanisms and road infrastructure have greatly reduced the number of road traffic casualties, decision makers still suffer sleepless nights as car accidents continue to be a significant cause of civilian death and injury. In the wake of such a disaster, we tend to hear that it is impossible to predict or prevent another accident, and the only solution offered is usually to close off that section of the road.

To reduce or, better yet, eliminate the road death toll, what is needed is a focused, effective response, and a concrete plan of action.

Recently, data collected from sensors installed throughout urban roads has begun to be used to provide some kind of solution. This big data is run through prediction models allowing municipal authorities to anticipate the location, time of day, and even the number of casualties of the next accident in the city. The success rate of this approach is impressive – in a 2018 survey of data we received from public databases containing information on road accidents in two large cities in Europe and the US, we found that we could predict the next accident with an accuracy of more than 80%.

Many cities are now trying to reduce the number of road accident casualties. They are seeking creative solutions that will enable them to map locations that are prone to accidents, and suggest changes that can be made to reduce the number and severity of accidents in those areas. The good news is that the infrastructure for such changes already exists – the growing trend of smart cities is the best thing to have happened to our personal safety, both on and off the roads. The more advanced cities in the world already have in place systems that collect data and establish databases (Big Data) from a variety of IoT sources located throughout the city – security cameras, traffic control cameras, smart urban lighting, traffic lights, water and sewerage systems, sensors at municipal parking lots, and air and water quality monitoring devices.

Today, municipal authorities understand that the large amount of information collected from these various devices and sensors can be used to reduce the occurrence of road accidents. Business Intelligence (BI) tools built in to advanced smart city management systems are able to exploit this information to reach conclusions that can be used to influence the public agenda, and support decision-making processes about the city’s operations and planning.

Smart city management systems support two operational scenarios: first, the day-to-day running of the city; and second, real-time support in emergency situations. In both cases, the management system absorbs large amounts of data enabling a better understanding of the operational situation, so that the required response can be formulated, and the relevant measures can be activated (either automatically or manually). The process is critical to a city’s real-time operations in emergency situations, or in situations in which a rapid, effective response is required. Analysis and implementation by rule base engine is carried out in real-time in parallel to the collection of data, enabling a rapid and accurate response to events such as flooding, accidents or fire. A truly advanced system can even automatically operate a response to a complex scenario, such as activating a public address system, dispatching teams to check an alert, diverting traffic to an alternative route, evacuating casualties or handling hazardous materials.

In addition to dealing with immediate operational needs, the same system serves municipal staff with extensive research that can be used to inform long-term operation and management decisions using Big Data analysis and forecasting tools to assist in the qualitative and quantitative understanding of events occurring in the city.

In both modes of operation, the system assists city managers in meeting pre-set goals as markers of success. These may include reducing the number of road accidents, public water consumption, response times or any other index – according to the city’s priorities. If a municipality has decided to deal with traffic accidents, the BI system enables a clearer view of the gap between its goals and the reality on the ground. It also facilitates research on the data that has been gathered, identifying the parameters that affect road accidents, and supporting the process of deciding how to handle the situation in a way that aligns with pre-determined priorities.

Another integral part of the city management system is its geographic information system (GIS) capability. This feature helps reduce road accidents and improve responses to emergency and other dangerous situations, by locating any data, whether a sensor feed or an event itself, in a specific geographic space, and carrying out a geographic analysis according to that location. So, for example, cameras located close to an operational or security event can be operated remotely to provide real-time surveillance, and also to provide a wider view of the surrounding area, to locate and capture broader situational factors. 

One area where this can be applied is in reducing the many accidents on busy main roads that involve electric bicycles. In order to predict the location of the next accident, we first pull statistics from the hundreds of municipal security cameras across the city, to find out which locations electric bikes tend to pass through the most. The next step is to take the images and analyze when the bikes are passing through a particular spot, how many there are, and at what speed. This information is then cross-referenced with data from the appropriate governmental or municipal authority about the number and types of accidents occurring in the city, to enable a better understanding of what causes accidents. The BI ANALYTICS capabilities of the system can then be used to test the effect any given factor, by changing a specific element and re-running the forecast model to evaluate and predict the effect of that change. For example, we can check how additional street lights being placed at a dangerous location could reduce the number of accidents in future.

Many municipal authorities today understand that city management systems that include command and control modules are a critical component for handling routine and emergency situations in the city, and an efficient and effective tool for gathering large amounts of data, analyzing it, and gaining insights into how to make improvements. When it comes to road accidents, anything that can be done to reduce the number of road accidents, decrease their severity or improve the efficiency of first responders, will clearly save lives and prevent more people being injured.


Recent Posts


Smart City | IoT | Command & Control | Management Platform | Homeland Security | Fiber Optics | Communication Infrastructure | Telecom Towers | Mega event | security | Airport security | Seaport security | Critical infrastructure security | Transportation security | Border protection | Prisons security | LTE | Cyber | Cyber | Threat Intelligence | Endpoint Security | MSSP | Emergency Management Security | First Responders | Tactical Defense | Communication | Mobile platform | Radio Coverage | In-Building Communication