Implementing Visual Knowledge for Flawless Design


VKFD Cover

Dr. Allen Ward wrote Visual Knowledge for Flawless Design more than 15 years ago for use in teaching the companies that he was working with… some of that work in cooperation with Michael Kennedy. After Allen’s death in 2004, TCC was founded by Michael and Brian to carry on that work teaching companies how to use visual knowledge.  In the 15 years since then, working with companies in a variety of industries, including aerospace, automotive, semiconductor,  and medical devices, we have made many significant improvements on the foundation originally laid by Allen.

In particular, to handle highly complex problems crossing different areas of expertise, we found that we needed to improve on the causal diagrams that Allen had proposed. That led initially to our Decision Map, which gave a visually clearer representation of the causal relationships.  However, we found that teams struggled to get from blank sheet to a good Decision Map.  So, we designed an intermediate tool, the Causal Map, along with a step-by-step methodology that would reliably get teams to a Decision Map that captured all the knowledge needed to make the complex decisions that they needed to make.

Causal Diagram without Circles
Causal Map without Relations


Similarly, we found teams struggled to come up with the right sets of Limit Curves and Trade-Off Curves.  Trying to find unitless ratios like the Reynolds numbers used in aerospace was not easy advice to follow. And building set-based charts that allowed teams to see their full multi-dimensional design spaces was difficult, if not impossible, with tools like Excel, LiveMath, or Mathcad (as Allen recommended at that time). It is very difficult for most engineers to imagine in advance what such charts should look like, so they have a hard time even formulating what they want until they see it. What they really needed was a tool that would allow them to visually navigate around in that multi-dimensional design space until they could see something useful… something that would help them make the decisions they needed to make.

Pressure Tank Charts


So, we built such a tool… that is now a key piece of our Success Assured® software!

To teach visual knowledge, Allen’s book uses an example of designing a pressurized tank that won’t have weld ruptures. 

Pressure Tank


He fleshes out a causal diagram and then identifies clusters within that for which a trade-off curve can be developed.

Causal Diagram without Circles


He then derives a formulation, reworking it to unitless ratios, and then shows how that trade-off chart can be used to guide decision-making. 

Pressure Tank Chart


We teach very similar, but with some very important refinements…

First, circling clusters that can be charted turns out to be visually awkward with even moderate complexity.  Those clusters will necessarily overlap with the neighboring clusters, creating tremendous visual complexity and confusion. Worse, as the real problem complexity rises, those clusters often get spread apart in the diagram, increasing that visual confusion such that the diagrams become unusable.  

Causal Diagram with Circles


To solve that, our Decision Maps use rectangular Relation shapes to represent the Trade-Off Charts that can be formulated between the Decisions (for which we use circular Decision shapes such that we can more easily attach information such as units of measure and characteristics such as being a Customer Interest).

Causal Map with Relations


Second, the + and - notations on the lines in Allen’s causal diagrams indicating “tends to make better” or “tends to make worse” actually tend to create confusion because, for many of the intermediate and root decisions, what is “good” or “bad” is unclear.  That can result in unnecessary debate on the proper directions of the indicators.  Similarly, the arrow heads indicating the causal direction also creates unnecessary debate because in reality, the cause-and-effect direction depends upon where you are in the decision-making process.  Consider F=ma … is it the force on the mass causing the acceleration?  Or is the required acceleration causing you to apply the necessary force to the mass?  Or is the increase in mass due to other decisions causing your acceleration to go down or your force to go up? Our problems are complex enough without adding extra things to debate or cause confusion.  We instead use a combination of arrows and +/- symbols to indicate how the values of the decisions actually move, without any implication of causal direction or any implication of good or bad.

Shuttle Areas of Expertise




Third, the Relation shapes between the Decision shapes simply capture what is known in whatever form it is known.  The person with the knowledge need not figure out how that knowledge will be best shown to the future decision-makers… that is left to the future decision-makers. Rather, the person with the knowledge on that portion of the Causal Map can simply capture what is known however the experts in that space are most accustomed to seeing it… making it easy to review and improve over time.  The tools then help the future decision-makers reuse that knowledge by showing it to them the way they want to see it.

Shuttle Causal Map


So, once a Decision Map has been constructed and the knowledge about the Relations captured, the decision makers can quickly use the Map to specify the multi-dimensional Trade-Off Charts that they want to see. They can slice-and-dice the design space in numerous ways allowing them to gain a deeper understanding of the multi-dimensional design space than any two- or three-dimensional chart could give alone.  They can learn how best to make the decision by visually exploring the multi-dimensional design space.

Shuttle Chart 1
Shuttle Chart 2


Once they work out the best way to make a set of decisions, they can capture the associated Trade-Off Charts and Trade-Off Solvers as a second layer of reusable knowledge built on the first layer of reusable knowledge (the Decision Map made up of Decision and Relation K-Briefs).  Those Maps, Charts, and Solvers can then be organized into a cohesive problem-solving or project-design K-Brief that tells the larger story, and forms the third layer of reusable knowledge.  Those three layers of reusable knowledge are a substantial upgrade to the visual Trade-Off Curve Sheets that Allen developed, and are far better able to handle the real-world complexity found in industries such as aerospace or medical devices.

KBrief LoRes


If you would like a demo of where we’ve taken Visual Knowledge for Flawless Design in the 15 years since Allen wrote the book, contact us now. We can show you Allen's simple model of a pressure tank… or we can show a bit more complex pressure tank model proposed by NASA, of the cost and aerodynamic optimization of the External Fuel Tank for the Space Shuttle.  Or if you would like to see an even more complex model, we can look at the mission design of an unmanned aircraft carrying IR tracking equipment.  Email us now!

Contact us now to learn more
© 2004-2018 Targeted Convergence Corporation