TourSim ABM

Update March 2014: It has come to my attention that in the last7 years since I made these Java applet models, that the security for in-browser Java applets has gotten much stricter. This is probably a good thing. However, this does mean that the links to the models below, which used to run in most browsers, no longer will, even if you add the site as a security exception. I’m taking a look into what would be required to update them, but it is entirely possible that they won’t be able to be updated to run in a browser. Such is the price of web security!

Update December 2011: This page will now serve as an ‘archive’ of my TourSim work. Creating and evaluating TourSim was a significant portion of my PhD dissertation. You can read about the development and evaluation of this ABM in the peer-reviewed journals listed below. I have several research directions that I am actively following based on my TourSim work, including a redevelopment of TourSim, and the use of ABM as a tool within a community development and visioning process. Please contact me if you are interested in either details on TourSim or on my current ABM and Geoweb work.

2011. Johnson, P. A., & Sieber, R. E. An agent-based approach to providing tourism planning support. Environment and Planning B. 38(3), 486-504.

2011. Johnson, P. A., & Sieber, R. E. Negotiating constraints to the adoption of agent-based modeling for tourism planning in Nova Scotia. Environment and Planning B. 38(2), 307-321.

2010. Johnson, P. A., & Sieber, R. E. An individual-based model of tourism dynamics. Tourism Analysis. 15(5), 517-530.

2009. Johnson, P. A., & Sieber, R. E. Agent-based modeling: A dynamic scenario planning approach to tourism PSS. In S. Geertman & J. Stillwell (Eds.), Planning Support Systems: Best Practices and New Methods. Berlin: Springer.


Thanks for taking a look at my TourSim model. You can run basic simulations in your web browser by selecting one of the links below. As of February 2009, the development of TourSim has reached a point of stability, so the model versions listed below are the most current available. I’ve started developing some more exciting applications of this technology as well as working on evaluating these existing versions. Enjoy!

In order to run these models

– You need to have pop-ups enabled on your browser

– Java must be enabled on your browser. If your version of Java is out-of date, you may be prompted to download the new version.

– You will be asked to click through a pop-up stating that this model was created with the educational version of AnyLogic modeling software.

1. April 30th, 2008 Adaptive Destinations with user control

I’ve made some major alterations to TourSim, both in the data that it relies on, and the types of experimentation it supports. I’m thinking that this is going to make TourSim much more usable for tourism planning, and begins to incorporate many of the ideas of complexity science (such as adaptation) into TourSim.

First, TourSim now uses tourist preference data from the 2004 Nova Scotia Tourist Exit Survey. This survey has a wider range of accommodation and activity options, and the types of categories represented relate much more intuitively to the types of tourism products available in Nova Scotia. Additionally, the number of responses included in the Tourist Exit Survey is considerably larger than the CTS and ITS I have previously been using. The Tourist Exit Survey also segments tourists based on generating market (Atlantic, Quebec, Ontario, Western Canada, New England, Other USA, and International). Each class of tourist has their own range of activity and accommodation preferences, and you can now see the percentage of each market that is arriving in Nova Scotia.

I’ve also improved the destination adaptation function. This is designed to represent destination development in response to high levels of visitation. Several steps are used to model this function:

1) Destination Capacity. Each destination has a maximum capacity for visits, based on occupancy data provided by a mandatory reporting program conducted by the Nova Scotia Department of Tourism, Culture, and Heritage. While this capacity varies considerably from season to season, this occupancy limit represents the maximum accommodation capacity if all accommodations are open.

2) Every month, the destination examines the number of tourists who have visited in that month. If the destination is at 80% of its capacity (this threshold is adjustable by the user), then the destination increases its capacity by 5% (this percentage again can be adjusted by the user).

3) Advertising: This adaptation function also works for destinations that don’t come close to their capacities. If a destination is below 30% of their capacity, the destination “advertises” and raises the likelihood that it will be randomly selected for evaluation by the tourist. Of course, this isn’t exactly how advertising works, but in the simplified world of TourSim, things are a bit different.

2. April 13th, 2008: Adaptive Destinations (click here to launch the model)

This model adds adaptive characteristics to destinations. Each destination has a representative level of capacity to accommodate tourists. Once a month, each destination checks to see if the number of tourists per day is within 80% of the destination capacity. If so, the destination adds 20% more capacity. In this manner, successful destinations grow to accommodate more tourists. I’ve added a toggle to turn this feature on and off, so you can test the difference that adaptation makes. Please note that without adaptation turned on, some destinations reach their capacity levels and effectively plateau. More refinements to come!

3. March 6th, 2008: Hotel Development in Baddeck (click here to launch the model)

I’ve now decided to start putting the date with each model revision, since adding a new title isn’t always possible. This version adds the ability to zoom in on Cape Breton for a closer look at how the tourist routes are developing, as well as 9 total destinations to view, and a few other bug fixes and refinements. The model now runs for 4 years before stopping automatically. As with the previous version, this model is designed to look at the difference that occurs when adding a seasonal (may to october) hotel at Baddeck.


Archive of older versions.

I’m just leaving these as a record of how TourSim has progressed.

1.Tourism Development Model (click here to launch the model!)

This model lets you manipulate average length of stay and annual growth rate. You can select to view results from a number of destinations and view dynamic charts. You can also look at the results of adding a hotel at Baddeck.

2. Tourist Port of Entry Scenario (click here to launch the model)

This is a basic scenario that lets users change the percentage of tourists entering at one of seven ports of entry (Yarmouth, Digby, Amherst, Tidnish, Pictou, Halifax Airport, and North Sydney). A dynamic bar chart shows the cumulative number of tourists that visit a large number of Nova Scotian destinations.

3. Port of Entry Comparison Model (click here to launch the model)

This model is a variation on the first port of entry scenario, but it allows you to capture results on the total number of tourists to visit Lunenburg, Baddeck, and Digby and then compare these results using a line graph. This version was designed as a way to make it easier to see what the effect of changing model parameters is, without exporting raw data to Excel (still a somewhat cumbersome process, and very difficult to do over the web, due to security concerns).

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Research on GIS, Open Data, and Mobile Technology, supervised by Peter A. Johnson at the University of Waterloo

%d bloggers like this: