Personalized day tour design for urban tourists with consideration to CO2 emissions

Abstract

The growing awareness of climate change worldwide has led the urban tourism market to focus on balancing tourist tailored experiences and CO₂ emissions. Therefore, designing personalized tourist routes with environmental pollution consideration is preferable in this context. This study proposes an evolution algorithm based on reinforcement learning (FSRL-HA) to design a personalized day tour route that simultaneously considers the utility of tourists and the carbon emission. We conducted a case study in Chengdu, Sichuan, China, to evaluate this algorithm's performance. The results indicate that the proposed algorithm outperforms selected baseline methods. Furthermore, the approach can provide more diverse route choices for different tourists, and an experiment was conducted to explore how tourist preferences affect tourist utilities.