Office of Research A Stimulated Annealing and Hill-Climbing Algorithm for the Traveling Tournament Problem

The Traveling Tournament Problem (TTP) schedules a double roundrobin tournament to minimize the total distance traveled by competing teams. It involves issues of feasibility and optimality and is challenging for both constraint programming and integer programming. This paper considers the TTP proposed in [4],[[12]. We divide the search space into two subspaces. Simulated Annealing (SA) is used to search a timetable space, while hill-climbing explores a team assignment space. The SA algorithm mutates timetables by means of conditional local jumps to find timetables which lead to better schedules. The hill-climbing algorithm is enhanced by pre-computation and dynamic cost updating and provides fast and efficient search. Experiments have shown that this hybrid approach give results on benchmark sets comparable to best known solutions. (