About Set-Based Games

This website is designed to be a fun way to explore the power of set-based analysis and decision-making. The best way to learn something that is a paradigm shift (like set-based is vs. point-based) is to apply both approaches to a few different examples, so that you get some real experience with the impact it can have.

To facilitate this, we have created a series of illuminating examples of increasing complexity. The goal is to keep the games simple enough that they don’t take too long, but complex enough that you can see how the many advantages would map to your own real-world problems (which are likely even more complex).

Start with the Tutorial that will explain each step, the basic concepts, and the pattern you’ll be following in all the rest, but is too simple to be an entertaining challenge. As you progress to the Advanced and Expert games, consider finding a few people to team up with such that you can work the games together. Some of the significant advantages of set-based analysis and decision-making over point-based is the superior collaboration — it is hard to experience that without actually trying to collaborate to solve a problem using each approach.

Each of these games can be played in just a few minutes if you just skim the knowledge and then take “educated guesses” and rely on the results from the built-in analyses to guide your iterative guesses. But if you invest a little more time, you can more closely simulate how you do your real work… at which point you’ll likely gain a lot more insight about the impact set-based could have on that real work. But you can also run through these quickly at first to see if it seems promising; and then replay the more advanced games more seriously, potentially even collaborating with some of your colleagues to better simulate your real work.

Credits

The idea for this site was inspired in part by the work of Kerga, Rossi, Taisch, Terzient on the use of "Serious Games" to teach Set-Based Design concepts. All the example content used here is borrowed from the Help K-Briefs of the Success Assured® software made by Targeted Convergence Corporation.

What is Set-Based Design?

Originally a term coined by Dr. Allen Ward from studying the product development processes at Toyota Motor Corporation. He described it as:

"it is an approach to design problems in which designers think and reason about sets of design alternatives. Over time, these sets are gradually narrowed as the designers eliminate inferior alternatives until they find a final solution. This approach differs from the common practice of making iterations (i.e., making several modifications of improvements in series) of one alternative until a satisfactory solution emerges."
Toyota, Concurrent Engineering, and Set-Based Design - Ward, Sobek, Cristiano, Liker

While not an engineering problem, one illustrative story to grasp the spirit of Set-Based Design is to imagine a team of people needing to schedule a meeting to discuss a future product. These are busy people and it is difficult to find a time that works for all concerned. Each person has their own set of constraints and limits that must be considered to find a meeting time that satisfies all variables simultaneously. In a point-based approach to picking a meeting time, the email exchange might look something like:

Person A: “Hey guys, I was thinking we could meet Fri at 8am to discuss our upcoming product. Would that work for all of you?”
Person D: “That would work for me.”
Person B: “Sorry. I’ll be out Fri. Vacation.”
Person C: “Yeah, I have a training class Fri. Could we meet Monday at 4pm instead?”
Person B: “I could do Monday at 4pm.”
Person D: “Monday works for me.”
Person A: “Darn. I’m booked all Monday with conference calls and Customer meetings.”
Person D: “Tues?”
Person A: “Nope.”
Person B: “Nope.”
Person C: “Nope.”

This team might eventually stumble upon a day and time that works with all member's constraints simultaneously by repeated Guess-and-Test loops, but it might take longer than it should and prove to be a frustrating experience.

Another option would be a converging, set-based approach. Each team member would first review their schedule for the week and identify a set of days where they would have at least some available time to attend the meeting.

When presented visually, it become easy to intuitively see what the feasible option is. All other weaker options can be quickly discarded from consideration. If the constraint sets are mathematically intersected, the feasible option becomes trivially obvious.

Then the analysis is further converged by each team member identifying the range of times they would be available for a meeting on the specified day.


Again, infeasible regions can easily be identified and rejected. By intersecting the set of constraints, the solution space identifies itself. No churn required.

Also, consider the case where there is no solution at all. Being able to detect no solution can save a company a lot of time and money thrashing around for a solution they will never find. If you have to fail, it is always better to "fail fast, fail cheap".

How do you play?

Imagine that you work for a company that designs and manufactures custom solutions for its customers. Your customers come to you with some set of requirements that may or may not be feasible and they ask you for a quotation. You need to analyze their requirements and design the best possible solution. If you cannot meet their requirements, then you may want to propose a few options that the customer can choose between.

At the start of the game, you will be given a requirements specification as part of a "Request for Proposal" (RFP) by the customer which may be somewhat ambiguous (as they tend to be in the real world). Your job is to formulate a proposal to the customer that consists of one or more feasible designs that they can choose from.

The game is played in two stages:
In the first stage, you are provided the kind of information you might have available in a conventional point-based environment. You use that point-based knowledge to devise a potential design. You can then “test” that design which will run a point-based analysis to determine the performance of that design vs. the requirements. Based on the results of that analysis, you can then devise alternative designs and test those. Ultimately, you’ll want to choose one or more of those designs to submit as your response to the customers’ RFP.
In the second stage, you are provided with set-based knowledge that gives you visibility to the sets that work and the sets that are infeasible, and then guided in how to use that to converge to the set (the portion of the design space) that you think would best satisfy the target customer.
Finally, we look at how your results from each stage compare, and then pose some questions for you to consider in comparing point-based vs. set-based.