Decomposition Creates Dysfunction

Systems Engineering techniques based on decomposition follow a Blind Men and the Elephant paradigm, where we assume (wrongly) that if a group of people thoroughly understand the components of a system, that they can also understand the system itself. To understand complexity, Systems Engineering is not enough, we must adopt Systems Thinking.

In this lecture, I cover how the very human decisions which led to hundreds of deaths and billions in losses in the GM Ignition Switch Recalls, and the Boeing 737 Max Groundings were due to the cultural and economic pressure to achieve revenue goals, and a lack of information or occasional recognition of seemingly unimportant information, which does not coalesce into systemic understanding. To understand systems, we must enable collective learning, but this is antithetical to the business models of the past century, which focus on divisions, departments, and componentry. (More info on GM and Boeing can be found here: Systemic Complexity (patrickhillberg.com))

I have had concerns for some time about the viability of autonomous vehicles (AVs) for passenger travel, and the week I recorded the lecture GM's Cruise Division pulled all of their vehicles off the road due to recurring safety concerns. I speak briefly on this here, as it seems to be a continuation of the GM and Boeing use cases.

G.M.’s Cruise Moved Fast in the Driverless Race. It Got Ugly. - The New York Times (nytimes.com).

Finally, I pose three hypotheticals, and would love to hear your response:

1. Is there a Limit to Growth in the number of safe and effective lines of code which can be embedded into a vehicle?
2. Should the AV manufacturers compete on the safest vehicle? (Safest to pedestrians?) If so, should any vehicle which is not the safest be allowed to operate?
3. Is it possible for individuals to add components into a system such that the system is impossible to comprehend, and if so, what is the likely result? What is the Limit to Growth in Complexity?

Additional Notes

After posting the above, I received a couple of comments which deserve more discussion than I can fit into a 1250-character LinkedIn Reply, so I'll add to them here.

The comments referred to INCOSE (the Internation Council of Systems Engineers) and their definition: What is Systems Engineering (incose.org),

"The System Engineering perspective is based on systems thinking. Systems thinking is a unique perspective on reality—a perspective that sharpens our awareness of wholes and how the parts within those wholes interrelate. When a system is considered as a combination of system elements, systems thinking acknowledges the primacy of the whole (system) and the primacy of the relation of the interrelationships of the system elements to the whole. Systems thinking occurs through discovery, learning, diagnosis, and dialog that lead to sensing, modeling, and talking about the real-world to better understand, define, and work with systems. A systems thinker knows how systems fit into the larger context of day-to-day life, how they behave, and how to manage them."

Honestly, I just don't see a definition of systems thinking there, and I am a reasonably knowledgeable person in this. I have an M.S, a Ph.D., and a faculty position in Systems Engineering, and the paragraph reads like a word salad. I feel like I should bring INCOSE into my teaching, but I'm stuck on that paragraph. How can I teach it to my students if I don't understand it?

Systems Engineering Book of Knowledge (SEBoK)

INCOSE is one of three stewards to another information repository, called the SEBoK, and it includes this: Systems Thinking - SEBoK (sebokwiki.org). TBH - following it is a bit of a rabbit hole, but it does provide references to additional authors, including Peter Senge and Donella Meadows.

The Fifth Discipline (Senge: 1999, 2006)

In my teaching, writing and speaking I reference The Fifth Discipline, by Peter Senge, of MIT. It has over two million copies sold (in comparison, INCOSE has twenty thousand members) and was identified by the Harvard Business Review as one of the seminal business books of the past 75 years. The book has 18 chapters, is over 500 pages long, and provides both a detailed and very readable build towards Senge's concept of Systems Thinking. It is assigned reading in my course. The following is from The Fifth Discipline - Wikipedia:

The Five Disciplines of what the book refers to as a "learning organization" discussed in the book are:

1. "Personal mastery is a discipline of continually clarifying and deepening our personal vision, of focusing our energies, of developing patience, and of seeing reality objectively."[2]
2. "Mental models are deeply ingrained assumptions, generalizations, or even pictures of images that influence how we understand the world and how we take action."[2]
3. "Building shared vision - a practice of unearthing shared pictures of the future that foster genuine commitment and enrollment rather than compliance."[2]
4. "Team learning starts with 'dialogue', the capacity of members of a team to suspend assumptions and enter into genuine 'thinking together'."[2]
5. "Systems thinking - The Fifth Discipline that integrates the other four."[2]

Senge describes extensively the role of what he refers to as "mental models," which he says are integral in order to "focus on the openness needed to unearth shortcomings" in perceptions. The book also focuses on "team learning" with the goal of developing "the skills of groups of people to look for the larger picture beyond individual perspectives."

System "Thinking" vs. "Engineering"

What is Systems Engineering (incose.org) never mentions these things, particularly the importance of collective learning. It never mentions the importance of feedback loops, archetypes, or organizational behavior. While I'm not about to be the person to define "Systems Thinking", I posit that we cannot understand a system without understanding Senge's Five Disciplines. These are not covered in the traditional Systems Engineering Vee model. (At about 11:00 in the lecture I state that my minimal definition of Systems Thinking must include feedback loops:

And shortly after I posit that Systems Thinking is a different dimension. I go on to say that Systems Engineering methods can help with complicated products, but complexity requires an understanding (the Senge definition) of Systems Thinking.

I'm slowly building lectures regarding Systems Thinking, based largely on the Senge definition. You can find these here: Systemic Complexity (patrickhillberg.com)