Time-Dependent Alternative Route Planning.

We present a new method for computing a set of alternative origin-to-destination routes in road networks with an underlying time-dependent metric. The resulting set is aggregated in the form of a time-dependent alternative graph and is characterized by minimum route overlap, small stretch factor, small size and low complexity. To our knowledge, this is the first work that deals with the time-dependent setting in the framework of alternative routes. Based on preprocessed minimum travel-time information between a small set of nodes and all other nodes in the graph, our algorithm carries out a collection phase for candidate alternative routes, followed by a pruning phase that cautiously discards uninteresting or low-quality routes from the candidate set. Our experimental evaluation on real time-dependent road networks demonstrates that the new algorithm performs much better (by one or two orders of magnitude) than existing baseline approaches. In particular, the entire alternative graph can be computed in less than 0.384sec for the road network of Germany, and in less than 1.24sec for that of Europe. Our approach provides also quick-and-dirty results of decent quality, in about 1/300 of the above mentioned query times for continental-size instances.