A fuzzy GRASP for the tourist trip design with clustered POIs

In tourist sector, expert and intelligent systems should perform at least two main tasks or services: point of interest recommendation and route generation. In this regard, the personalized electronic tourist guide, generally implemented on hand-held device, such as mobile applications or in web. These tools must work like an expert and intelligent system that perform the services mentioned above then they should need low computation effort. In this paper we focus on the route generation based on scores of the points of interest and the distance or time between them. We consider a new extension of the Tourist Trip Design Problem, named Tourist Trip Design Problem with Clustered Points of Interest, where points of interest are grouped in clusters representing different types of attraction sites. Moreover, minimum/maximum limits are imposed on the number of points of interest belonging to the each clusters that are visited in the same route. Since it is a novel problem, we generate two sets of instances in order to evaluate the accuracy of our solution approach. A Fuzzy GRASP (Greedy Randomized Adaptive Search Procedure), in which both distance based and score based evaluation criteria are used to guide the candidates selection in the construction phase is proposed. The results provided by our heuristic are compared with those obtained by solving the MIP formulation. Computational results carried out on real and real-like instances show the effectiveness and efficiency of the proposed approach and its suitability to be part of a Personalized Electronic Tourist Guide in hand-held devices. (C) 2019 Elsevier Ltd. All rights reserved.