Tourism combination forecasting with swarm intelligence

Combination forecasting is an effective method for improving the accuracy of tourism demand. This study proposes an innovative combination strategy based on a multi-objective swarm intelligence optimization algorithm and, for the first time, examines whether and how this algorithm can enhance the pe...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Annals of tourism research Jg. 111; S. 103932
Hauptverfasser: Li, Hengyun, Guo, Honggang, Wang, Jianzhou, Wang, Yong, Wu, Chunying
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Ltd 01.03.2025
Schlagworte:
Combination forecasts
Multi-objective optimization
Swarm intelligence optimization algorithm
Tourism demand forecasting
Multi-objective optimization
Combination forecasts
Swarm intelligence optimization algorithm
Tourism demand forecasting
ISSN:0160-7383
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Combination forecasting is an effective method for improving the accuracy of tourism demand. This study proposes an innovative combination strategy based on a multi-objective swarm intelligence optimization algorithm and, for the first time, examines whether and how this algorithm can enhance the performance of tourism demand combination forecasting. An empirical study conducted under several scenarios demonstrates that the proposed combination strategy enhances the interaction among single forecasts, leading to improved forecast accuracy and stability compared with traditional combination methods. The model remained effective even during the COVID-19 pandemic. The findings have a positive impact on predictive research, offering new insights and methodologies for tourism demand modeling. •This study forecasts daily and weekly tourism demand for three tourism destinations.•A novel combination method based on multi-objective swarm intelligence is proposed.•The proposed method can enhance both forecast accuracy and stability.•The proposed method can improve forecast accuracy even in turbulent period.
ISSN:0160-7383
DOI:10.1016/j.annals.2025.103932