“I study traffic patterns,” special agent Ethan Hunt described his fake job at the beginning of “Mission Impossible 3.” He went on to explain what fascinated him about traffic: “You hit the brakes for a second, just tap them on the freeway, you can literally track the ripple effect of that action across a two-hundred-mile stretch of road, because traffic has a memory. It's amazing. It's like a living organism.
Ethan had a point. At the very least, traffic does resemble a living organism in that it can suffer from congestion. But why? The US Department of Transportation divides the causes into two types. Recurring congestion forms because there are too many cars on the road at once. Non-recurring congestion occurs due to unforeseen circumstances such as accidents, bad weather, or road works.
Out of the two, non-recurring congestion is much more difficult to predict and manage. We expect a traffic jam at 9 AM on weekdays because we realize that many people drive to work around that time. Accidents, on the other hand, happen at random. Similarly, bad weather, although possible to forecast, has unpredictable impacts on traffic. It can bring about puddles, icy spots, wind-borne obstacles, and more. Only roadworks can be planned and announced in advance unless they result from bad weather or accidents.
For businesses, traffic jams are always detrimental, but especially so when they are random. According to “Road Traffic Congestion: A Concise Guide,” traffic slowdowns inflate business costs by increasing travel times, fuel consumption, the likelihood of collisions, and even health care expenses due to excess pollution. But if a traffic jam is unexpected, it can drive up transportation costs by 50 to 250%!
The costs of getting stuck in traffic encouraged automobile drivers throughout history to devise ways of predicting and avoiding slowdowns. To that end, we listen to traffic reports on the air, communicate over Citizens Band radio, and read electronic highway signs. More recently, we also use smartphone applications such as Google Maps.
Google Maps remains one of the most popular traffic navigation apps. On its website, it boasts 97% accuracy for its travel time estimates. Is it an impressive number? It depends. Unfortunately, Google neither elaborates on the claim nor provides any source to support it. Therefore we can only speculate. 97% can mean that for every one hundred trips, three predictions are entirely inaccurate. It can also mean that time estimates are incorrect for three out of every one hundred kilometers. If that were the case, a driver could be stuck indefinitely somewhere along those three kilometers without even affecting the accuracy ratio. So we should approach the 97% figure with a grain of salt.
A more interesting question is how does Google arrive at its estimates? Their blog post explains that Google collects historical traffic data and supplies them to a machine-learning algorithm. Then, the algorithm uses the data to infer future traffic patterns. Certainly, however, such an algorithm is only capable of predicting recurring congestion. In other words, the kind of traffic that occurs consistently, like the one at 9 AM on weekdays. Random events, by definition, can not be predicted by analyzing the past.
What scale are we talking about for those random events? It turns out they can account for 60% of all traffic congestion. Accidents make up 25 to 35% of the whole, bad weather is responsible for 15%, and work zones for 10%.
Besides being random, these factors have something else in common. They all involve specific actors. Police officers, health care workers, firefighters, road managers, to name just a few, all come into play when the unexpected happens. In fact, it is not an overstatement to say that they are on the front lines of up to 50% of all traffic incidents. And they use their own communication channels that regular drivers have no access to.
At Clurgo, we have developed a novel technological solution that fills a gap in traffic tracking. It taps into primary sources and collects the most reliable information about unexpected traffic events in real-time.
Our solution is called KPD. It is a centralized database platform with a uniform interface that connects road managers, public servants, and any actors involved in traffic management. The platform enables them to map and track traffic incidents in real-time. Whenever a random event occurs, a person with access to KPD can add it to the database with a few taps. From that point on, the event is visible to every user. They can then pass the information to the radio, the web, and even to Google Maps.
Of course, we would all prefer Ethan Hunt to study traffic for us and deliver the information to our ears in his one-of-a-kind soothing voice. But before he shows up, we can offer you the next best thing: KPD – an award-winning, real-time traffic incident tracker, the first solution of its kind in the country.