Simulation and Models in Product Development
Modeling Is Not As New As You May Think.
Models are not new, and neither models in the employ of product development. Product development has always had some basis in discovery and always will. If everything had such a high degree of certainty, likely the product or endeavor has already been done. Developing new things ceaselessly brings questions. To be effective, we want to answer these questions as quickly and as certainly as possible. Furthermore, in the course of answering the questions, we often discover more prudent questions to ask. In the olden days, we would create the product, learn from it and modify until we asymptotically approach success or perfection. Model based development is not that new though. It was not ushered in when computers entered our post industrialization era. In fact, anybody that lives in North Carolina would likely know the story of the Wright Brothers (yes I know they were from Ohio – a fact my wife will never let me forget). The brothers had begun building their understanding of wing performance based upon the work of Otto Lilienthal. By the way, Otto Lilienthal died as a result of injuries sustained in a glider accident. No doubt this influenced how the Wright brothers preceded in their product discovery exercises.
Kites and Gliders.
In the brothers initial work with kites and gliders, they noticed a difference between the theoretical calculation and the actual performance of the model or kite. That is the kite did not react as predicted based upon calculation. They knew then that something was amiss. The doubts they had required exploration. An example of the experiments they performed can be found here. To that end the brothers decided to measure and find the real answers for themselves. The brothers set about creating a wind tunnel and tools and techniques that would allow them to understand on a smaller and more controllable scale those things that matter to achieve flight. This arrangement made it possible for the brothers to not only arrive at the understanding of lift and drag, but made it possible to critique a variety of wing geometries to determine the best solution, all without the risk to life and limb. This was in the year 1901.
This was not meant to be a story about those who were part of the discovery of flight, but their story reflects a long ago instance where models were used to understand how things work. My guess is that as far back as can be measured we would see similar instances. Perhaps the first spears and nets followed a similarly process of discovery.
Today, with high powered computers, and the ability to create mathematical models of the components, not just physically but also from the performance perspective we are able to learn much more and faster. We can model entire vehicle systems before we begin to produce the first prototype parts. We can gather environmental data from the vehicles through instrumentation of existing vehicles while undergoing typical or stressful stimulation. We can even gather some vehicle use information from telemetry systems on a product or customer vehicle, allowing us to modify or adapt future vehicle systems based upon real data from use.
Physical World and The Product
That is not to say that model based development is not without difficulty, or provides an instantaneous of understanding. We must have a way to generate the models, in this context we are referring to an understanding of the physical world in which the product will inhabit. That requires an understanding of the variations possible both in system configuration as well as external stimulation or collection of stimuli. Those attributes that are connected to the desired outcome, we must understand so we can accurately predict the outcome. Below find a graphic that shows how this works[i]:
Success is not due to randomly generating models via a plethora of assumptions. It is the learning and understanding of the key variables (in context of the Wright’s work lift, drag, weight and wing shape and angle of attack) in the environment then mimicking those in some manner of simulation (the Wright’s wind tunnel and models of wings). We will want to compare the real world reaction to our model predictions and understand and resolve the difference, and we will want to do learn as quickly and clearly as possible. To accomplish this goal, we will measure these real world attributes. Our understanding of the real world will make it possible for us to design the product that will perform how we want it to in the context of the environment in which it will reside.
Learn more about simulation and prototype parts.
[i] Pries, K., & Quigley, J. (2011). Scrum project management. Boca Raton, FL: Taylor and Francis Group, page 134.