This paper develops an ant colony optimization approach to the orienteering problem, a general version of the well-known traveling salesman problem with many relevant applications in industry. Based on mainstream ant colony ideas, an unusual sequenced local search and a distance based penalty function are added which result in a method that is convincingly shown to be the best heuristic published for this problem class. Results on 67 test problems show that the ant colony method performs as well or better than all other methods from the literature in all cases and does so at very modest computational cost. Furthermore, the ant colony method is insensitive to seed, problem instance, problem size and degree of constraint.