Simulation as a Development Method
Key Points:
- The state of the simulation is the state of understanding.
- Simulation pulls from all resources, makes the pieces interact, and illuminates gaps and next steps.
- The Goal: Make the prototype the confirmation, not the experiment.
The idea of a digital doppelgänger is not new, and there are practitioners of this art form, but very likely not enough. Human nature tends to drive optimism and skipped steps. The digital doppelgänger, or evolving system/process simulator (hereafter simulator), helps enforce a level of reality and discipline around recognizing gaps and defining next steps. It gathers the various pieces together and makes them interact—the environment in which the system operates, the physics, the controls, and the economics. This allows us to get a handle on the key questions: How will the system work under different conditions? How do we engineer and control it to work optimally? How do we maximize the value proposition? We are compelled to chase the gaps to make progress, rather than allowed to work only on what is comfortable and familiar.
The ideal simulator is able to make an impact immediately using first principles and available data. In this way, gaps can be identified and sensitivities determined that then guide subsequent information gathering, whether from literature, computational modeling, or testing. The incorporation of test data is of paramount importance. A relatively small number of real-world tests can be used to tune and validate simulators. These simulators can then run orders of magnitude more virtual tests to determine sensitivities, performance envelopes, control strategies, and much of what ultimately needs to be sorted out for the actual product. With targeted testing comes the opportunity to build test rigs specifically designed to produce the data needed, in sufficient quantity and quality, far faster and more efficiently than could be done with a prototype.
The act of building the simulation is an opportunity to differentiate between what is known and where the gaps remain. Resources can then be focused on those gaps. Simulations of necessity tend to start simple, but they should be built with the hooks necessary to incorporate everything needed to quantify the value proposition. They should also bring business, product, and marketing development into much earlier and more intimate contact with engineering and technology development. The same simulator can be used by all of these groups to advance their individual objectives while driving the success of the whole.
One special advantage of Vox Cassandrae is that the full range of useful skills, training, and experience is integrated. The test rigs and DOEs are fit for purpose because they are specified by the same minds that will also process the data, understand how it was generated, build the models, and incorporate those models into simulation. The improved simulation then reveals the next steps. It becomes a virtuous cycle of progress.
I have seen this methodology accelerate development of the recipe for success far more effectively than skipping steps and hoping the prototype will reveal the answers. It can do so at lower cost, with greater agility, and in some cases with important secondary benefits such as improved safety. At the intersection of these various tools and methods, brought together by simulation, is the clearest picture of the state of development. As the simulator matures, so does the technology. When the important physics, behaviors, controls, and economics are sufficiently understood and integrated, much of the intellectual work required to build the real system has already been completed.
At that point, prototype construction becomes increasingly a commodity exercise—and a confirmation of the technology rather than the experiment that discovers it.
— Cassandra
William Bell Scott - Iron and Coal (1855-1860)
