We consider vehicle routing problems in the context of the Air Force operational problem of routing unmanned aerial vehicles from base locations to various reconnaissance sites. The unmanned aerial vehicle routing problem requires consideration of heterogeneous vehicles, vehicle endurance limits, time windows, and time walls for some of the sites requiring coverage, site priorities, and asymmetric travel distances. We propose a general architecture for operational research problems, specified for vehicle routing problems, that encourages object-oriented programming and code reuse. We create an instance of this architecture for the unmanned aerial vehicle routing problem and describe the components of this architecture to include the general user interface created for the operational users of the system. We employ route building heuristics and tabu search in a symbiotic fashion to provide a user-defined level-of-effort solver interface. Empirical tests of solution algorithms parameterized for solution speed reveal reasonable solution quality is attained.
[1]
Marius M. Solomon,et al.
Algorithms for the Vehicle Routing and Scheduling Problems with Time Window Constraints
,
1987,
Oper. Res..
[2]
Gilbert Laporte,et al.
The vehicle routing problem: An overview of exact and approximate algorithms
,
1992
.
[3]
Fred W. Glover,et al.
Future paths for integer programming and links to artificial intelligence
,
1986,
Comput. Oper. Res..
[4]
William B. Carlton,et al.
Dynamic Routing of Unmanned Aerial Vehicles Using Reactive Tabu Search
,
2001
.
[5]
Russell L. Ackoff,et al.
Resurrecting the Future of Operational Research
,
1979
.