Why 98% data accuracy isn’t enough for your traffic solution

By: Miovision Team | Oct 21, 2021
98% Data Accuracy - Miovision

The safety of citizens on roadways and at intersections is a top priority for governments, and drives decision-making at every stage of traffic infrastructure implementation. From strategic planning, to procurement, installation, and management, ensuring public safety comes down to having accurate and timely data to make the right decisions at the right times. So when your team is deciding on new traffic technology for your city, data accuracy rates play a critical role in ensuring public safety and shortlisting vendors.

Many vendors will claim that their traffic technology is 98% accurate, which sounds impressive and implies their solution is highly reliable. But this go-to marketing claim about traffic technology performance doesn’t tell the whole story.

Claims of 98% data accuracy aren’t enough. Ask to see the vendor’s traffic technology in action, with real-world data that shows post-implementation results – don’t just take their word for it. What’s really important is what happens in the other 2%. What do they do to ensure public safety when the system isn’t able to collect accurate data to control signal phasing information?

How to dig deeper on accuracy

Look for results captured in the real world

For many vendors, 98% data accuracy is often achieved in lab settings, where conditions are ideal. If tested in the real world, under variable conditions such as time of day and inclement weather, that percentage could be very different. Claims of 98% data accuracy mean nothing if a system fails when it’s needed most, like at dusk or in inclement weather.

Ask to see videos of how the technology performs at the most challenging intersections when there’s a glare, when it’s snowing, at night, etc. Understanding the methodology that was used and the conditions in which stats were calculated will help you cut through the marketing claims to see the numbers that matter most.

Look for more than just TERL certification

Florida’s Traffic Engineering Research Lab (TERL) has developed standardized testing and certification of traffic control signals and devices. TERL certification is becoming an expectation — and in some cases, a requirement — when it comes to selling traffic control devices or intelligent transportation systems to State Departments of Transportation. But in the government’s own words, their standards represent the “minimum specifications” for certification of a solution.

The Traffic Engineering Research Laboratory (TERL) exists as a joint research facility between the Florida Department of Transportation and the Florida Agricultural and Mechanical University-Florida State University College of Engineering. Its mission is to conduct applied research, standards and specifications development and support the evaluation and testing of Intelligent Transportation Systems (ITS) and traffic engineering technologies.

Check for evaluations that go above and beyond TERL standards of accuracy and safety. Look for ones that assess detection accuracy for all types of transportation (including pedestrians and cyclists) and the accuracy of turning movement counts (TMCs) of vehicles, cyclists, and pedestrians in crowded urban centres.

Look for 24/7 support to ensure accurate and timely data

When it comes to public safety, there’s no room for downtime and no time to wait on customer service. Look for a vendor that offers 24/7 support that will step in to help you look at the data, interpret the issues, and provide a resolution.

Know the details around what’s included in customer support and training post-implementation, and make sure there are no hidden costs or loopholes in service offerings. Choose a vendor that will ensure you’re getting accurate and timely data to make the right decisions quickly and easily. And when the system isn’t working as intended, that there are fail-safes in place to reduce the risk of collisions.

Look no further than Miovision

At Miovision, “good enough” isn’t in our vocabulary. Our testing is conducted in real-world conditions and has a TERL accuracy rate of 99.2% plus. But we go above and beyond – proactively refining our models and algorithms with machine learning for continuous improvement of accuracy. And once it’s installed, it only gets better. This is why Miovision achieves even higher standards when it comes to data-driven insights and safety.

In addition to TERL detection accuracy certification, we’ve added two of our own standards by which we measure our performance:

  • A qualification that evaluates detection of all modes of transportation
  • A qualification that evaluates the detection of turning movements through study-based vehicle counts

And because transparency is at the heart of everything we do, we’ll show you real-world performance of our technology at challenging intersections. See first-hand how our products improve traffic flow and safety in cities all around the world, regardless of the time of day or weather. Our technology helps to optimize traffic flow, provides auditable video and results, and has built-in safety measures that automatically revert to equal light time durations on all sides in zero visibility conditions.

Miovision’s always-on support is included in our SaaS subscriptions – so you can rest assured knowing there are no extra fees or hidden costs when reaching out for help. We’re there when you need us and we’ll see the problem through to resolution. All fixes are pushed out across every implementation of Miovision technology, so that everyone benefits from the updates.

Want to learn more? We’re ready to answer all your questions and show you the data.

Get in touch for a demo.