Afbeelding
Tourism and recreation are of great importance to tourist destinations. However, too much tourism leads to crowding and possibly nuisance for the community and landscape. In order to be able to manage and anticipate crowds, there is a great desire among policymakers, entrepreneurs and industries to be able to predict this better. Is it possible to predict where and when it might be busy on the basis of big data? Our program managers Jasper Heslinga and Harm IJben are investigating this question in our project “Predictive models for insight into tourism pressures”.
Experiment with predictions
During an expert session on forecasting models, 6 data experts exchanged experiences and knowledge in order to be able to further develop the models. The experts discussed:
- A handy literature review on forecasting models presented as well as experiments (by Noël Middelhoek of the European Tourism Futures)
- Findings about making predictions based on GPS data (by Jolene Cijsouw of Hogeschool Zeeland)
- First version of a forecasting model using AirBnB data (by Janneke Arntzen and Oswald Dijkstra from DataFryslân)
- Experiences with predictions and prognoses (by Tanja Fedorova and Joris Klingen from the Municipality of Amsterdam)
Hopeful experiments, but next steps needed
The literature study provides a good overview of what a model is, which methods are available and when they can be used under which conditions. This "quick scan" provides a useful overview to make a good start with forecasting models. The experiments show that research based on time series does not work well enough, but that regression models yield better results. There is a lot of potential in forecasting models, but more extensive testing with other data and combining with other methods needs to be done. The overarching conclusions and lessons learned from this project will be published shortly.
Collaboration around tourist data
This project is part of the Data & Development Lab, in which CELTH has joined forces with Statistics Netherlands and NBTC Holland Marketing to experiment with data and information for the leisure and tourism sector. For this project, there is collaboration between the European Tourism Futures Institute (ETFI) and the Knowledge Center for Coastal Tourism. Thereby, there is intensive collaboration with the data experts of DataFryslân and the research group 'Data Science' of HZ University of Applied Sciences. In this way, the knowledge of tourism and leisure is combined with the technical knowledge of modeling.
