GraphHopper Directions API News June

The GraphHopper Directions API is continually improved and we’ll keep you up to date with the newsletter, today the first time publicly available as blog post.

We added more features to the Route Optimization API

• Often there are more customers than the available vehicle fleet can serve. In this case, you can now assign priorities to customers that must be served to differentiate them from those that can be served later or omitted at all.
• You can integrate load dependent transport costs, i.e. we consider that it is usually more expensive in terms of energy consumption to drive a full truck load, i.e. unnecessary kilometers with full truck loads can be avoided
• Many auto generated clients for the different programming languages like C#, php, ruby and python are now available

The Geocoding API now allows you to specify an external provider like Nominatim or OpenCageData using the same API and no need for additional contracts. We’ll add more soon, if you need one which is not yet there, please ask for addition here.

The Matrix API has a more memory efficient Java client.

A more flexible routing is now available in the Routing API:

• You can now create round trips with the Routing API using ch.disable=true and algorithm=round_trip, plus some more optional parameters
• An often requested feature is the turn restriction support, which is now possible (example) and
• additionally alternative routes can be used to provide users a nice choice
• When navigating with motor vehicles it can be important to avoid to turn at destination or via points, this is now possible with the heading parameters
• A snapped_waypoints field in the JSON allows you to calculate the distances from road to the provided destination or via points e.g. to consider more time when planning a delivery route
• You can now calculate the shortest not only the fastest path too, via weighting=shortest (and ch.disable=true), but keep in mind that only very few use cases require the shortest path.

Last but not least a new ‘hike’ vehicle is available for all APIs, this profile is more suitable for tourism navigation e.g. prefers hiking tours. Instead the ‘foot’ profile prefers shorter routes which is more suiteable for inner city pedestrian navigation