What is next for Model-Based Design?

I have drawn the software design “V” roughly 1e04 times.(1) Over time, the scope of what is in the V for Model-Based Design has increased. Sometimes I think there should be a Model-Based Design equivalent to Moore’s Law (4); perhaps the “delta V”.(5)

In the arms of the V

Often when asked “what’s next?”, deeper is the answer. Improve a product, improve the process, increase (or decrease) the outputs; the less obvious answer is what can be brought into the embrace of the V.

As the scope of Model-Based Design expands, it has done so in a technology first, workflow second hodgepodge.(6) Uniquely what we are now seeing is the embrace of model based systems engineering which focuses on workflow first and tools second.

What is a system?

Definitions of systems are generally unsatisfying as there is not consensus on what comprises a system; however the definition I like best is…

Open and Closed Systems: A system is commonly defined as a group of interacting units or elements that have a common purpose. … The boundaries of open systems, because they interact with other systems or environments, are more flexible than those of closed systems, which are rigid, and largely impenetrable

Come together, right now, over???

I doubt the Beatles were thinking of software systems when they wrote this song (7) but it gets to the heart of why systems are important. They give people a place to “come together.” No model is, or should be, an island.(8) But how do we come together? The answer is through abstraction.

How does the centipede walk? (9)

The objective of a systems integration environment is to provide an abstracted language so that software, hardware, controls, physical modelers… can all exchange “information” without having to understand the heart of each other’s domain.

At a minimum, a system acts as both a “living” ICD document and as an ICD verification tool. When working in the metadata rich Model-Based Design environment a system integration tool can do much more.

Used well, metadata is an asset, used poorly it can sink your ship.(10) Use of a systems integration tool provides a natural “organizational” layer on top of the metadata. System interactions, data dependencies, system wide requirement coverage all can be accessed from within a system level model.

The Worm Ouroboros: Model-Based Design or Model Based System Engineering

I’ve been asked “What is the boundary between MBD and MBSE”? Right now where it lies is an open question (11) and not at present a fruitful question. Five years ago there was a clear answer, and five years from now we will again have clarity, but right now we are in a land rush (12) where developers can explore new ideas and stake out new domains. What is next? A better understanding of what is.

Footnotes

  1. I came to this number using a Fermi Estimation methodology.
    Years in MBD Community ~ 20, Customer per year ~ 50, Drawings per customer ~ 10
    Years mentoring in MBD ~ 10, mentored per year 4, drawings per mentored ~ 20 (2)
    tot = 20 * 50 * 10 + 10 * 4 * 20 = 10,800. (3)
  2. In the end, the mentored end up drawing it and having their own way of talking about the design V.
  3. The Fermi estimate may make you think that the 800 from mentoring is not important; however, if I count the number of times the people I taught in turn draw this, then I may hit another order of magnitude.
  4. It always seemed that Moore’s law should be called “Moore’s extrapolation.”
  5. You could think of each software release as an instantaneous acceleration.
  6. Technology first, workflow second: a tool for solving a specific problem is developed; how that tool fits into the overall workflow comes later (along with refinements to the tool).
  7. And I doubt it was an endorsement of open source software.
  8. John Donne poem “No man is an Island” : (with apologies)

No man is an island entire of itself; every man
is a piece of the continent, a part of the main;

No block is a model entire of itself; every block
is a peace of a subsystem, a call from the main();

  1. The Centipede’s Dilemma, is a something organizations hit when they start to examine how they do something. Much like the Coyote who can run on air until they look down.
  2. Though I suppose in the case of the titanic they had the data to go by, it a structural problem.
  3. The image of an “open system” was chosen to lay the ground work for this section.
  4. For those of you less familiar with United States history, a “land rush” refers to a time period in American history when settlers could “ride and claim” a parcel of land. When I was a child this was taught as a positive moment in history; as an adult, it is easy to see the casualties and costs to Native Americans.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.