On the tour planning problem

• Published in: Annals of operations research Vol. 192; no. 1; pp. 67 - 86
• Main Authors: Zhu, Chenbo, Hu, J. Q., Wang, Fengchun, Xu, Yifan, Cao, Rongzeng
• Format: Journal Article
• Language: English
• Published: Boston Springer US 01.01.2012; Springer Science + Business Media; Springer; Springer Nature B.V
• Subjects: Advisory services; Algorithms; Approximation; Approximation theory; Business and Management; Catering; Combinatorics; Computation; Customers; Electronic commerce; Heuristic; Hotels & motels; Laboratories; Linear programming; Mathematical analysis; Mathematical models; Methods; Mixed integer; Operations research; Operations Research/Decision Theory; Orienteering; Planning; Recommender systems; Studies; Technology application; Theory of Computation; Tourism; Tours; Travel; Travel agencies; Travel industry; Websites; China; Local search; Team orienteering problem; Mixed integer linear programming; Tour planning problem; Heuristic
• ISSN: 0254-5330; 1572-9338
• DOI: 10.1007/s10479-010-0763-5

Increasingly, tourists are planning trips by themselves using the vast amount of information available on the Web. However, they still expect and want trip plan advisory services. In this paper, we study the tour planning problem in which our goal is to design a tour trip with the most desirable sites, subject to various budget and time constraints. We first establish a framework for this problem, and then formulate it as a mixed integer linear programming problem. However, except when the size of the problem is small, say, with less than 20–30 sites, it is computationally infeasible to solve the mixed-integer linear programming problem. Therefore, we propose a heuristic method based on local search ideas. The method is efficient and provides good approximation solutions. Numerical results are provided to validate the method. We also apply our method to the team orienteering problem, a special case of the tour planning problem which has been considered in the literature, and compare our method with other existing methods. Our numerical results show that our method produces very good approximation solutions with relatively small computational efforts comparing with other existing methods.