-1%20(1).png)
Published by Valeriya Murzyna on
Moore’s Law is far from dead. When Gordon E. Moore predicted in 1965 that our computer processing capacity would double every two years, he probably wasn’t thinking about cell phones and smart cars.
And he certainly wasn’t thinking about Big Data for traffic management. But today’s traffic managers are surrounded by a wealth of data, and the volume of that data continues to grow exponentially. Jumping into the deep end of this massive data pool can be intimidating, but it’s a worthwhile endeavor. Big Data has multiple innovative applications in traffic control and management.
Reduced Congestion and Better Parking
Congestion has long been a fundamental challenge in traffic management, and the rapid growth of urban centers has only made it a more pernicious problem. Limited parking in city centers often exacerbates congestion, as cars circling to find parking spots clog up roads.
Conventional efforts to alleviate congestion might involve road tubes or manual counting, but these methods are extremely limited–they provide no data on vehicle classification, speed, direction or timing. Moreover, they tend to focus on the most heavily traveled roads, so less traveled thoroughfares get ignored.
Instead, traffic managers are increasingly turning to a host of other data sources, most notably live traffic video and other real-time monitoring devices. These devices can generate a wealth of data in very little time. Of course, this data can be used to make immediate traffic control decisions, but it can also be archived and aggregated into a true Big Data source.
By strategically using Big Data, traffic managers can reap benefits immediately and over time.
- Travel-time reduction (short term): Identifying peak travel times and analyzing travel time reliability can provide the necessary information to help drivers make better choices about when to travel, and which routes to take. Real-time data can also be used to inform drivers about potential delays due to heavy congestion or accidents.
- Demand-based strategies (middle term): The root causes of traffic congestion can be better identified and addressed. For example, vehicle classification data might reveal that congestion is caused by trucks entering and leaving a distribution center.
- Infrastructure investments (long term): Historical data can help quantify the value of goods and jobs that move along specific transit routes, which can then be prioritized for additional infrastructure investments.
.webp?width=759&height=493&name=AdobeStock_70020384%20(1).webp)
Less Expensive, More Holistic Transport Planning
High-quality data is absolutely essential for transport planning. Yet traffic modellers have traditionally used analog sources like the Household Transport Surveys (HTS), which is both expensive and difficult to administer. Although the HTS offers a strong link between traffic models and reality, it also provides a relatively limited picture of transportation; sample sizes are often quite small (less than 2% of the target population), which isn’t enough to illustrate transport trends.
To complement this information, traffic managers can now glean Big Data from a variety of relatively new sources, which include the following:
- Mobile phone data: This includes data generated by both the Mobile Network Operator (MNO) and the Global Navigation Satellite System (GNSS).
- Smart cards: These provide relatively reliable data regarding public transportation use.
- Geocoded social media records: A relatively new data source, social media and other apps can provide reliable data on people’s point-to-point movements.
- GPS navigation: Vehicle-based GPS navigation offers point-to-point travel data, although it doesn’t provide data for the movement of individual people.
- Commercial fleets: To generate additional revenue, long-haul and regional fleet companies share anonymized data with various solution providers.
Transport planners can pair this information with data generated from traffic cameras and sensors. For example, traffic video analytics might include data regarding vehicle counting and classification; speed, direction and timing. Nowadays, you can also get “origin-destination” traffic data from cameras and video analytics systems, e.g., using license plate recognition to reidentify vehicles using footage from different cameras.
By bringing together Big Data from such a wide range of sources, traffic modelers can make much more accurate, comprehensive predictions and forecasts. Moreover, the cost of conducting surveys and other data collection can be significantly reduced.
.png?width=800&height=488&name=pasted%20image%200%20(10).png)
More Equitable Transportation Access
That same cell phone data can also help traffic managers ensure that transportation access is equitable and truly serves each neighborhood’s unique population. Ensuring equity in transportation can be quite a difficult task, so many traffic managers are enthusiastic about using Big Data to address this challenge.
We mentioned that HTS surveys are expensive and time consuming. They also often yield biased results, reflecting only the habits, needs, and preferences of the most vocal communities–or at least the self-selecting group that has the time and desire to complete a long survey. On the contrary, the number of operating mobile phones reached almost 15 billion worldwide in 2021, and studies show that the demographics of mobile-phone owners are usually quite close to those of the general population in terms of age, race, education, and other factors.
The impact of Big Data gleaned from mobile phone users can support transportation equity in multiple ways:
- Identifying ways to optimize public transit routes for neighborhoods with low car ownership, in the context of residents’ shopping and commuting habits.
- Analyzing truck routes to better understand which neighborhoods and roads are disproportionately impacted by noise and pollution.
- Implementing safety measures for pedestrians and construction workers in high-speed corridors.
- Communicating with all residents who may be affected by traffic and infrastructure changes.
One excellent outcome of achieving better transportation equity is that resources tend to be allocated more efficiently across the transit system: resources essentially get “right-sized” for each community’s actual transportation needs.
The key to harnessing Big Data is to start with the right analytics tools. GoodVision provides a suite of video analytics software that works with your existing cameras to deliver real-time data on a host of critical metrics.
Watch our webinar below to find out more on how GoodVision can help Traffic Management. You will learn how to use digitally collected traffic data for various traffic behaviour analyses you need for traffic modelling, such as Vehicle speed, Gap-acceptance, Saturation flows, Occupancy and dwell time analysis.
