An interview with TrafficCast's Nick Kiernan

Traffic abilities in GPS is a hot topic, but how does it work? To learn more about the technology, services, and future, Miss Direction interviews TrafficCast's Nick Kiernan.

Of all the new capabilities being introduced on portable navigation devices (PNDs) and GPS-enabled cell phones, real-time traffic is probably one of the most-wanted and hottest features today. It makes sense. After all, traffic plays a huge role during your travels, whether you're just making your everyday commute to work or heading off on a holiday.

All that said, just how useful are the traffic capabilities on GPS devices? Are there limitations? Where does the data come from? To help you (and me) better understand the technology and types of services, I talked with Nick Kiernan, Vice President of Business Development for TrafficCast, a traffic service provider. Check out our conversation below to learn more about the company, traffic data in PNDs and cell phones, and the future of GPS.

Can you tell me a little bit about the history of TrafficCast? How and when did it start?
TrafficCast was established in 2002, emerging out of a successful traffic and transportation consulting business, TranSmart Technologies, which was founded by Dr. Connie Li and Dr. Bin Ran. Dr. Li and her colleagues saw a need for more precise, accurate traffic prediction products and began developing models and analysis algorithms based on their backgrounds in transportation engineering. The founders worked to ensure that these models were deeply rooted in the science of traffic, incorporating elements such as Traffic Flow Theory, which has enabled TrafficCast to produce solutions that fill-in data gaps that have previously plagued the industry.

How have you seen the space change over the years?
The traffic industry has changed dramatically over the years as new technologies and distribution channels develop. In the 1970s and early 1980s, we saw companies such as Metro and Shadow Traffic emerge as helicopters, traffic cameras and the all important "tipster" became increasingly popular. These companies would take the data they received, aggregate it, and produce it for broadcast on local television and radio stations. In the 1990s, with the growth of the Web, we saw companies such as Traffic.com emerge to provide graphic information for new media and television. This is also when we began to see speed sensors utilized in the roadways to measure traffic flow.

Today, companies such as TrafficCast are taking this to another level by incorporating information from emerging technologies into our traffic models. Incidents, traffic flow, weather, GPS probe data and other sources are now used in complex models to accurately predict traffic across a number of national markets. The channels we have to distribute this information have also expanded to include GPS devices, mobile phones, Web portals and traditional media.

What types of products and services does TrafficCast offer?
TrafficCast's core product is called Dynaflow, which actually is broken down into two versions. Dynaflow 1.0 and Dynaflow 2.0. Dynaflow 1.0 models the impact of a number of dynamic conditions to produce "real-time" traffic flow information. This product aggregates more than 350 data sources and uses our patented models to provide our customers and partners with accurate traffic information they then deliver to consumers. The key to this product is that the models it uses are able to fill-in the "gaps" that exist with current sensor and probe data. Because of this, the product produces data for primary, secondary and tertiary routes. This is the product companies such as Yahoo are using to provide real-time traffic information for their map services.

Dynaflow 2.0 focuses on long-term predictive traffic information. Using historical traffic patterns and information such as upcoming events and weather, this product provides 48-hour forecasts of road speeds to help with trip planning, etc. This long-term predictive information can be very valuable to travelers as well as fleet companies whose drivers may have trips lasting more than two days.

It's important to recognize that traffic information by its nature - even what's called "real-time" is predictive. Even road sensors involve analytics to derive speeds from the data points. Our models extend road coverage by expanding data integration and analysis. We also work with partners to provide products like RouteCast, which combines traffic and weather to offer consumers a personal traffic report based on the flow on their desired route and the weather's impact on this flow. We continue to develop additional products and applications that will truly bring actionable traffic data to consumers through a number of channels.

What sources/technology do you use to collect the data?
As mentioned above, our models incorporate more than 350 data sources which are analyzed to produce our accurate real-time and predictive traffic information. This data includes multiple sources of incidents, real-time and archival traffic sensor flow, real-time and archival GPS probe data, weather and events. Once we have aggregated the information provided by our data partners, our patented analytic models go to work to process all of this data and provide accurate traffic information for more than 100 national markets.

How do you differ from other traffic providers?
The foundation of our company and the approach we take to traffic information differs greatly from other traffic providers. TrafficCast was founded by distinguished traffic engineers who come to the table with a very scientific approach to traffic information and analysis. Because of this, the models and algorithms we use take many elements into consideration that go beyond the basic aggregation of data. Our models incorporate factors such as Traffic Flow Theory which enable analysis that can accurately predict how a single incident may impact traffic flow at different periods of time or on surrounding roadways. We are the only traffic data company holding patents for the software and scientific models we use.

What are some of the biggest challenges for you? Whether it be collecting the data, distributing the information, the limitations of the technology, or something else?
Perhaps one of the biggest challenges is addressing the consumption of data. Particularly in the mobile environment, the user interface involves critical safety and systems integration issues. It's more than just posting icons on maps; delivering actionable information requires software and application development that drivers can rely on without compromising the safe operation of their vehicle.

Do you get any data back on usage patterns? If so, what are people looking at? What type of traffic data is most important to them?
We don't have any data on the usage patterns of traffic information in regard to how people are getting their traffic information. However, when looking at the PND industry, we do know that uptake of traffic data subscriptions has been very low, with most reports indicating it remains in the single digits.

This is probably why we have seen companies, such as Navigon and Garmin, go to a subscription-free traffic model with their products. Along with the additional cost, which consumers may be resistant to, we feel that the content needs to be accurate and actionable, and presented in a way that is easy to digest. That is, what good does it do to know there is an accident on your route if you are already sitting in the traffic jam caused by the accident?

We are finding that people want traffic data that will help them avoid congestion and get home or to their child's soccer game quickest. This means that it doesn't even need to be tied to a map, something people may not think about. This could be as simple as a daily text alert at 5 p.m. telling someone whether to take route A, B, or C home based on the current traffic. We all know how to get home--we probably know five ways to get home. What we don't know on any given day is which way will be less congested.

What are some of the most requested features from your customers?
Traffic information is all about the value of time. All the data in the world won't necessarily reduce traffic congestion, but with the right context it can help people make decisions that save time or just reduce stress. Everyone wants a feature that "avoids traffic" but perhaps what the market really demands is accurate and reliable information about individual routes and the alternatives.

What types of features are you working on to implement in the future?
We are hard at work developing a number of new technologies we feel will make our data even better and also expand upon the services we can offer to our customers and consumers. One of the most exciting projects we are working on surrounds something we call Dynamic Traffic Assignment (DTA). This is something we feel could be the 'killer app" when fully functional.

The premise of this technology is that individual vehicles would be routed around congestion or traffic incidents in different ways. Think of the fact that with most of today's products, if there is an accident and you have 30 drivers with PNDs in their car alerting them of the incident, they will all most likely receive the same alternate route around the incident. So now you have 30 people clogging up another route and essentially the congestion has just been transferred to another roadway. With DTA, the individual drivers would be directed on various alternate routes in order to avoid transferring the congestion caused by the incident. In the ultimate example, everyone from a suburban neighborhood heading into the same vicinity of the city for work would be assigned various routes in the morning to spread traffic over different routes to reduce traffic congestion.

This is just one of the technologies we are working on. We are also exploring the best delivery methods for traffic information, such as the text alerts I mentioned above, how to incorporate two-way communications into our models, bringing a connected/community application to commuters to share traffic information, etc.

Where do you see the future of GPS? Will GPS-enabled cell phones and smartphones take over standalone PNDs? Or will there be a place for both?
We really see the GPS market becoming part of the converged mobile device market, so it is highly likely that the future of GPS may lie in GPS-enabled mobile phones. This isn't going to happen overnight, but as more GPS-enabled phones ship, more users will expect to have navigation and traffic applications on these devices, just as they now expect music players and cameras. The key here is that whichever way this goes, be it carriers and cell phone manufacturers developing navigation applications, or traditional PND manufacturers entering the mobile space, they must concentrate on the user experience and doing it right.

Everyone involved needs to be cautious and not jump head-first into throwing GPS applications on devices if it can't be done right. It is challenging to provide the experience a PND gives you on a smart phone because of things such as screen size or processing power. With traffic, this may be a little easier because as mentioned before, we don't necessarily feel that traffic information has to always be tied to a map. It can be a standalone application for consumers to use based on finding the least congested route from point A to B.

We do feel that the mobile/converged device industry will be the key area of growth for navigation and traffic applications. This is one of the main reasons we brought on our new CEO Neal Campbell, who has an extensive background in the mobile industry with Motorola. We feel that his expertise positions our company well as the shift to converged mobile devices continues. That being said, the standalone PND probably won't die any time soon. Dedicated devices typically perform better at their given task than a converged device, and you will always have consumers who appreciate the deeper experience they receive from a PND.

About the author

Bonnie Cha is chief correspondent for Crave, covering every kind of tech toy imaginable (with a special obsession for robots and Star Wars-related stuff). When she's not scoping out stories, you can find her checking out live music or surfing in the chilly waters of Northern California.

 

ARTICLE DISCUSSION

Conversation powered by Livefyre

Don't Miss
Hot Products
Trending on CNET

Hot on CNET

CNET's giving away a 3D printer

Enter for a chance to win* the MakerBot Replicator 3D Printer and all the supplies you need to get started.