Hybrid Genetic and Simulated Annealing Algorithm for Vehicle Routing Problem in Recommendation of Tour Route in Days

Abstract

Indonesia has a variety of amazing tourist attractions, making it difficult for tourists to choose which attractions suit their preferences. For vacation, people spend several days traveling to optimize their preferences, such as time, cost, and rating. Previous research has addressed the problem of recommending tourist routes for multi-day visits by analogizing it to solving the Travelling Salesman Problem (TSP), but this analogy only produces one route. Thus, the route must be trimmed based on time constraints per day. This causes the daily route to be suboptimal. Therefore, this research proposes a new method that analogizes this problem with finding a solution to the Vehicle Routing Problem (VRP) using Hybrid Genetic and Simulated Annealing (HGSA). By incorporating Degree of Interest (DOI) as a user preference, HGSA can recommend optimal routes based on individual needs. The recommender system we built is a framework. The framework can be applied to various types of datasets from all over the world. However, in this research, we use data on tourist attractions and hotels in Yogyakarta. Multi-Attribute Utility Theory (MAUT) is one of the methods used in multi-criteria decision-making. This research considers MAUT as a fitness value that evaluates various alternative routes generated based on several attributes or criteria. The experimental results demonstrate that HGSA outperforms other algorithms, achieving a fitness value of 0.7927, compared to the Firefly Algorithm (FA), Grey Wolf Optimizer (GWO), and Particle Swarm Optimization (PSO), which had fitness values of 0.7461, 0.7474, and 0.7445, respectively.