How Google Maps Calculates Traffic Jams

Recently, Berlin artist Simon Weckert ‘hacked‘ Google Maps by dragging a cart full of 99 iPhones running the app around the city. He practically deceived Google Maps into thinking there was an actual traffic jam, when in fact there was none.

Though it’s a clever trick, it raises important questions about how Google Maps keeps track of the data it collects in ordinary traffic conditions. If the stunt pulled off by Weckert is able to fool Google Maps’ system, can you always be certain about the app’s accuracy? From what we know about the app’s algorithm and Google’s responses to Weckert’s stunt, the answer is apparently: yes, mostly.

How Google Maps displays traffic jams

Google Maps continuously refreshes based on anonymously-tracked user data, traffic sensors and satellite data to ensure the most accurate traffic conditions are being displayed.

You may have seen streets coloured green, orange or red. Green means the traffic is normal but orange and red indicate slow-downs. They show up when users in navigation mode are travelling slower-then-usual. Once other drivers using Google Maps are able to wade through at normal speeds, however, Google will gradually remove the slow-down status for the area even if others are still slowly driving.

According to 9to5Google’s report on Weckert’s stunt, it took just a single car driving past him at normal speed to undo the traffic jam status which was caused by his cart of iPhones. Having said so, apparently Google Maps ignores when several users are at a standstill. Weckert’s wagon had to be moving for Google Maps to see the traffic jam despite the fact that nearly 100 phones were in navigation mode and not moving. It’s unclear how long a vehicle should be idling before it’s ignored; Google Maps will obviously notice if several users were stopped on a highway and reported that as a slow-down. However, if a group of people parked in a garage just had their navigation mode turned on, you probably wouldn’t see any traffic jams in the area.

Google Maps can tell the difference between a car, motorcycle and other vehicles (well, atleast in some countries). Even if a cheeky scooter weaves its way through a traffic jam, Google Maps could, in such a case, recognise the smaller vehicle’s movement and not register it as smooth flow of traffic.

Accidents, construction and traffic incidents

Google Maps also updates with traffic reports based on user feedback apart from parallel, anonymised location and movement tracking to keep conditions updated.

Accidents, construction sites, speed traps and other alerts can all be reported in the app. The more users reporting an alert makes it show faster on the map for others and help them avoid slow-downs to find better routes. Here’s how to report traffic incidents and slow-downs if you’re unfamiliar:

  • In Google Maps’ navigation mode, tap the “+” icon.
  • Tap the icon for the type of incident you want to report and follow on-screen instructions to finalise submission.

The app will begin adding your report to the map and you can undo the process during a very short timeframe.

Calculating reroutes and ETA

Google compares the speed of its users with speed limit information to calculate the Estimated Time of Arrival (ETA).

Google Maps will also cross-reference conditions reported by users and other traffic data with your current speed to adjust your arrival times if it notices a slow-down or other incidents along the route. Once the app determines the severity and length of a slow-down, it alerts users whose routes are affected and suggests alternative directions (reroutes).

To be noted that Google Maps’ suggested reroutes aren’t always perfect though. The algorithm can only respond to latest user data and guess, though can’t predict with certainty. Having said so, you can be confident Google Maps is giving you the latest information it has.

Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

To Top