Ask-the-Expert On-Demand - Implementing a Digital Twin for the Monopoly Board Game Using SAS® Viya®
Watch this Ask the Expert session to explore the use of digital twins to better understand the specifics of real-world environments.
Watch the webinar
You will learn more about:
- How the powerful SAS Viya platform creates Monte Carlo Simulations to create digital twins.
- Tips and tricks for using SAS data step, procedures, arrays, formats and more.
- How to combine the digital twin with SAS Visual Analytics to make your results more actionable.
The questions from the Q&A segment held at the end of the webinar are listed below and the slides from the webinar are attached.
Q&A
Does this only work for the Monopoly Board Game or also for complex business environments?
Yes, it also works for complex business scenarios. That’s one of the advantages we leverage in many custom implementation projects. We use this approach for “What-If” scenarios, such as making different decisions with a risk portfolio in banking or with various choices in a transportation network. The range of applications is very broad. I use the Monopoly board game as an example because it effectively illustrates what can be achieved and applied in these contexts.
What is the runtime of such a simulation with 10000 iterations?
The runtime that you saw in my game was with 1,000 iterations. You observed that this was running between one and two seconds. The SAS data step might get a little bit more comprehensive, if you consider profit calcualtions but will still run in a few seconds event for 100000s of iterations. If you then start to study in detail the balance and profitability of the fields, you will need to transpose a large number of columns. So, the simulation itself will still run in seconds, but you may want to put some thought into the data preparation piece—how to arrange the data nicely—which can take more than 10 seconds, but still remains quite efficient.
Do I need special programming skills to implement such a digital twin?
If you're a SAS programmer who is comfortable using a SAS data step with formats, arrays, and maybe some macro language, you're good. If you have SAS programming skills and experience working with SAS, you should be able to build your first digital twin.
Can you also do statistics on the significance of the differences between the yields of the fields? For example, in the error bars in the plot.
The answer is yes, you can do this. For example, you can achieve this in SAS Visual Analytics by running the analysis. However, I would like to clarify what is possible from a technical perspective. Still, I would be cautious about overinterpreting the statistical results because you are working in a simulation scenario. For instance, if you are looking at p-values or similar measures, simply increasing the number of simulations could quickly yield statistically significant results. However, calculating error bars, such as standard deviations, is certainly a reasonable way to examine the data. Yes, this is possible in both SAS procedures and in SAS Visual Analytics.
What makes this example different from say a simulation model?
I would say it is a simulation model. When I first published this 18 years ago, I simply referred to it as my Monte Carlo simulation for the Monopoly board game. Having seen many conferences and organizations discuss what constitutes a digital twin, I would argue that what we are doing here is a digital twin. It allows you to mirror a real-world situation digitally, which can be used to gain deeper insights. If someone prefers to call it a simulation model, that's fine with me. However, it is also a digital twin because it mirrors the Monopoly board game in a digital way and lets you experiment with it.
The most important question of all! Which colour properties should I buy?!
The question is whether you should focus solely on which colour properties to buy. I’m a big fan of looking at the outcomes—the results I showed you. Consider which fields have the highest visit probability and highest average balance, because this might not be limited to just one colour group; it could be individual fields. It may be wise to buy all properties of one colour group, but it could also be a good strategy to pick, for example, field numbers 12, 25, and 37 to get a better mix. Once you’ve run this analysis—and when I share the code, you’ll be able to run it yourself—you can also aggregate the results by colour group and get overall values for each, providing a clear overview. I’d like to emphasize that it’s not only about average balance. For some, a higher average balance for a field may seem attractive, but if the variability is also very high, it’s a riskier investment. In that case, it might be better to opt for a slightly lower balance with less variability and lower risk.
Recommended Resources
SAS Communities Articles "Simulating the Monopoly Board Game, Digital Twin"
- Monopoly #1 - Basic Example for the Visit Frequency of the Fields
- Monopoly #2 - Consider "Go-to-Jail" for the Visit Frequency of the Fields
- Monopoly #3 - Create a visualization dashboard in SAS Visual Analytics
- Monopoly #4 - Consider Chance Cards and Community Chest Cards for the Visit Frequency
- Monopoly #5 - Calculate the Profit per Field and per Player
Youtube Video: Studying Complex Systems – Simulating the Monopoly Board Game
Github Repo: Chapter 26, 27 and 28
SAS Press Book: Applying Data Science: Business Case Studies Using SAS®, Case Study 8
Please see additional resources in the attached slide deck.
Want more tips? Be sure to subscribe to the Ask the Expert board to receive follow up Q&A, slides and recordings from other SAS Ask the Expert webinars.
