The agent-based transport model SIMBA MOBi is applied by Swiss Federal Railways to plan future rail infrastructure, timetables, and service concepts. It reacts to changes in demographics (ageing, migration, household location), transport supply, and policy. We present an extension to model tourist behaviour. It should react to a change in tourist accommodation (e.g., a new hotel), in tourist attractions (e.g., a new museum) and in the tourist population (e.g., a change in the countries of origin of tourists in Switzerland) and enable a more accurate analysis of foreign tourists in train stations. The model is built in three steps. First, we generate agents, each with a country of origin and a tourist accommodation (hotels and supplementary accommodations such as holiday homes, collective accommodation and campsites). This part is based on data from the Swiss Federal Statistical Office (overnight stays in 2019) and Openstreetmap (location of accommodation as of today). Second, we generate a main transport mode for each agent. The estimation of the choice model is based on data from the Tourism Monitoring Switzerland 2017. Third, we generate activity plans for the agents and define a destination choice model.

1. 1
Antonin Danalet
STRC, Monte Verità, May 11, 2023
Modelling
tourists

2. Modelling foreign tourists in Switzerland.
2
Background Data Structure of the
model
Discussion

3. SIMBA MOBi: agent-based transport model at SBB
– Already integrates arrival/departure trips of tourists
– Border crossing traffic (rail and road)
– Traffic generated by airports (rail and road)
– Mobility of foreign tourists within Switzerland:
– For now, no behavioural model
3
Trips by non-Swiss tourists:
about 2% of total rail demand, but much more on
touristic lines.
1. Background.

4. 4
What happens when a site
development takes place?
Possible application:
New hotel.

5. 5
What happens when, e.g., a new
large museum opens?
Possible application:
New hotspot.

6. 6
It will change the choice of transport mode.
Possible application:
Change in the origin country of tourists.
42%
72%
58%
28%
European tourists
Non-European tourists
Use the train Don't use the train

7. 7
What is the percentage of tourists
among visitors?
Possible application:
Impact of tourists in train stations.

8. Holiday
homes
Campsites
Supplementary
accommodation
(PASTA)
Data.
8
Tourist
accommodation
statistics 2019
Tourism
Monitor
Switzerland
2017 Openstreet
map
Passenger
survey
Hotels
(HESTA)
Hotels
Collective
accommodation
Holiday
homes
Collective
accommodation
Campsites

9. Year of reference.
9
2017 2018 2019 2020 2021 2023
2022
If possible, including
Tourism
accommodation
statistics
2020 & 2021 not
representative for long
term planning
Swiss Travel System
tickets (sales data)
(or possibly 2019)
Tourism Monitor
Switzerland
New edition of the
Tourism Monittzerland,
available in 09.2024
Modelling tourists, Antonin Danalet

10. Structure of the model.
10
– Distribution of tourists in
accommodations
– Available variables:
– Country of origin
– Classification of the hotel
– Localisation
What is missing:
– Age
– Gender
– Duration of stay
Discrete choice model with:
– Country of origin
– Localisation of accommodation
(urban/rural & region)
– Type of accommodation,
classification of the hotel
– Gender?
– Age?
– Duration of stay in
accommodation and in
Switzerland?
Activity choice using MOBI Plans
– Attractivity:
– Number of visitors in hotspots
– Hypotheses
– Activity duration, departure time:
– Starting with leisure behaviour
of the Swiss resident
population
– Then adjustments?
What is missing:
– Distance distribution
Calibration: Passenger survey data
Discrete choice model to be
developed
Calibration:
Distribution of Swiss Travel System
tickets in the tourist population by
country of origin
Tourist population
Public transport
subscription
Main transport mode
Calibration & integration
in SIMBA MOBi

11. Tourist population.
11
Number of overnight stays (FSO)
• By commune (if 3+ hotels in the commune)
• By region/canton
Number of hotels and other accommodations (OSM)
including (partially) capacity/number of beds

12. 12
Hotels in Lausanne from Openstreetmap, including size.
Modelling tourists, Antonin Danalet

13. 13
Holiday homes in Switzerland from Openstreetmap.
Modelling tourists, Antonin Danalet

14. 14
Holiday homes in Verbier from Openstreetmap.
Modelling tourists, Antonin Danalet

15. 15
Holiday homes in Bern from Openstreetmap.
Modelling tourists, Antonin Danalet

16. 16
Main transport mode in Switzerland.

17. – Country of residence:
Continental Europe
– Type of accommodation:
Campsites, more stars in hotels
– Urban-rural typology of commune
of accommodation:
Staying in rural areas
17
– Country of residence:
UK, America, Australasia, Asia
– Region where the accommodation
is: Staying in Berner Oberland
Choice of train as main transport mode influenced by…

18. Activity plans: 20 «tourists» in Interlaken.
18

19. 19
1000 «Tourists» in Interlaken.
Destination choice.

20. 20

21. 21
Danke, merci
& grazie.