Spatio-temporal Networks: Modeling, Storing, and Querying Temporally-Detailed Roadmaps - Space-Time Integration in Geography and GIScience - page 88

the problem of determining the shortest travel time path between the University of
Minnesota and the MSP International Airport during the interval from 7:00 AM
to 10:30 AM. Figure 6.8 a shows two different routes between the University and
the Airport. The 35 W route is preferred outside rush-hour, whereas the route
via Hiawatha Avenue is preferred during rush-hour (i.e., 7:00 AM-9:30 AM) (see
Fig. 6.8 b). Thus, the ALSP route collection may be a set of two routes (one over
35 W and one over Hiawatha Avenue) and their corresponding time intervals.
Computing ALSP is computationally challenging for three reasons. First, many
links in the network may violate the property of first-in-first-out (FIFO) behavior.
This is illustrated in Table 6.6 , which shows the flight schedule for Delta airways
Fig. 6.8 Sample query for spatial big data. (Best in color). ( a ) Different routes between University
and Airport (Google Maps). ( b ) Preferred route varies over time. ( c ) Total travel time of candidate
paths between University and Airport (Color figure online)