A constraint programming approach for multi-objective tourist trip design problem with mandatory visits: A case study for İzmir Turkey


Işık E. E., Ayyıldız E., Taşkın A.

JOURNAL OF PROJECT MANAGEMENT, cilt.9, sa.1, ss.61-72, 2024 (ESCI) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 9 Sayı: 1
  • Basım Tarihi: 2024
  • Doi Numarası: 10.5267/j.jpm.2023.8.003
  • Dergi Adı: JOURNAL OF PROJECT MANAGEMENT
  • Derginin Tarandığı İndeksler: Emerging Sources Citation Index (ESCI), Scopus
  • Sayfa Sayıları: ss.61-72
  • Yıldız Teknik Üniversitesi Adresli: Evet

Özet

The Orienteering Problem (OP) is an optimization problem that finds the locations and routes that will return the highest profit/benefit, starting from the initial location of the traveler/vehicle, visiting these locations, and ending with the starting location of the tour within a given time or distance limit. There is no obligation to visit all locations in the problem structure. OP has many real-life applications, such as staff routing and disaster relief routing. In this study, OP with Time Windows (OPTW), an extension of OP, is discussed with hotel selection and mandatory visits. Although the main objective of OPTW is profit maximization, it is also essential to minimize the total travel time to complete the tour efficiently. For this reason, we consider the OPTW as a multi-objective problem. In the problem considered here, it is assumed that the profit/benefit, travel time between locations, service period, and time interval that each location can be visited are determined to be known. Within the scope of the study, first, a Mixed Integer Programming (MIP) model is prepared for the problem. Since the proposed mathematical model does not provide solutions in a reasonable time for large networks, the problem is solved by a Constraint Programming (CP) approach. Attractive tourist points of interest for Izmir, one of Turkey's major tourist cities, are determined, and the proposed method is applied to the real-life problem. The problem is modeled as Multi-Objective OPTW with MIP and CP and solved. Also, sensitivity analysis is performed by considering two different scenarios.