Transport Problems (Mar 2022)

DETERMINATION OF OPTIMISED PICK-UP AND DROP-OFF LOCATIONS IN TRANSPORT ROUTING – A COST DISTANCE APPROACH

  • Armin HAHN,
  • Wiard FRÜHLING,
  • Jan SCHLÜTER

DOI
https://doi.org/10.20858/tp.2022.17.1.02
Journal volume & issue
Vol. 17, no. 1
pp. 17 – 28

Abstract

Read online

With the emergence of dynamic passenger transport systems, such as demand-responsive transport (DRT) and ride-sharing without predetermined stop locations as used for static bus routes, accurate routing for these flexible door-to-door transport services is needed. Routing between two addresses requires the assignment of addresses to suitable, so-called snapping points as reference points on the road network. Therefore, many conventional routing machines use perpendicular distance to identify the nearest point on the road network. However, this technique tends to produce inaccurate results if the access to a building is not reachable from the road segment with the shortest perpendicular distance. We provide a novel approach to identify the access to buildings (paths) based on remote sensing data to obtain more reasonable stop locations for passenger transport. Multispectral images, OpenStreetMap data, and light detection and ranging (LiDAR) data were used to perform a cost distance analysis based on vegetation cover, building footprints, and the slope of the terrain to identify such optimised stop locations. We assumed that the access to buildings on the shortest route to the building’s entrance consists of little vegetation cover and minimal slope of the terrain; furthermore, the calculated path should not cross building footprints. Thus, snapping points on the road network can be determined based on the most likely path between a building and the road network. We validated our results based on a predetermined ideal snapping area considering different weightings for the parameters slope, vegetation, and building footprints. The results were compared with a conventional routing machine that uses perpendicular distance. This routing machine shows a validation rate of 81.4%, whereas the validation rate of our presented approach is as high as 90.3%. This new approach provides increased accuracy and better comfort for flexible passenger transport systems.

Keywords