The bi-objective prize collecting traveling backpacker problem for planning flight itineraries

da Costa, Calvin Rodrigues,Nascimento, Mariá Cristina Vasconcelos
DOI: https://doi.org/10.1007/s13042-024-02228-x
2024-06-14
International Journal of Machine Learning and Cybernetics
Abstract:Low-cost airlines have become increasingly popular in Europe as they offer affordable air travel for tourists on a budget who want to visit different countries. However, deciding on the places to visit can be challenging for travelers, especially when considering the cost of air tickets and the points of interest they wish to visit. This paper proposes the Bi-Objective Prize-Collecting Traveling Backpacker Problem (BO-PCTBP) to support this decision-making. The BO-PCTBP simultaneously minimizes air ticket costs and maximizes the sum of collected prizes assigned to the destinations. The time-varying availability and costs of the arcs between pairs of airports differentiate this problem from existing related routing problems. Besides the mathematical formulation, we present an innovative hybridization of the evolutionary multi-objective optimization algorithm Nondominated Sorting Genetic Algorithm II (NSGA-II). We conducted experiments using literature data and a new set of instances introduced in this paper, compiled from real-world low-cost airline data. The results indicate a desirable diversity in the Pareto approximations to enable distinct options for the traveler.
computer science, artificial intelligence
What problem does this paper attempt to address?