Cost-Based attraction recommendation for tour operators under stochastic demand

Abstract

Attraction recommendation is a key functionality offered by tour operators. The main stakeholders of attraction recommendations include tourists and tour operators. The former use recommendations to make travel decisions, and the latter manage recommendations for their own benefits. Most existing attraction recommendation methods focus on providing recommendations that best match tourists’ preferences, yet overlook the benefits of tour operators. To address this gap, we conduct a two-phase study that focuses on cost-based attraction recommendations under stochastic tourist demand from the perspective of tour operators. In the first phase, we obtain preliminary recommendation solutions that best match tourists’ topic preferences. Then, with the consideration of cost factors, a stochastic programming model with a joint chance constraint is proposed to refine the preliminary recommendation solutions in the second phase, and a tractable model based upon Sample Average Approximation (SAA) method is further presented. To assess the performance of the proposed method, comprehensive experiments are conducted with both simulated instances and real-world data. The results indicate that the proposed optimization model can significantly reduce tour operators’ recommendation cost while maintaining a high service level and tourist satisfaction.