Multi-attraction, hourly tourism demand forecasting

Forecasting tourism demand for multiple tourist attractions on an hourly basis provides impor-tant insights for effective and efficient management, such as staffing and resource optimization. However, existing forecasting models are not well equipped to hand the hourly data, which is dynamic and nonlinear. This study develops an improved, artificial intelligent-based model, known as Correlated Time Series oriented Long Short-Term Memory with Attention Mecha-nism, to solve this problem. The validity of the model is verified through a forecasting exercise for 77 attractions in Beijing, China. The results show that our model significantly outperforms the baseline models. The study advances the tourism demand forecasting literature and offers practical implications for resource optimization while enhancing staff and customer satisfaction. (c) 2021 Elsevier Ltd. All rights reserved.