I have been down in Virginia this week for some training and had some time to talk with some really smart people, something that I always enjoy. One of the topics that invariably pops up in the greater D.C. area is government contracts and the administration thereof. It all started with discussing how difficult it is to give initial estimates for work that will somehow, hopefully be within whatever the tolerance is of whatever the government’s independent estimators come up with.
These discussions are always fun to be a part of, especially when you are working with people from different industries. One came from a background of bidding as a prime to sub work, and touted his experience with LPTA contracting, while the other works in an engineering firm where the complexity of solutions runs contrary to the cost-only provisions of LPTA. Hearing the words complexity and solution in the same sentence always piques my interest, so I immediately started asking about how they manage to develop accurate estimates where uncertainty reigns.
The traditional means of reducing risk is to conduct planning in a predictive life-cycle project, or to conduct a spike in an agile organization. The issue is that funding and resources may not be available to a project manager who is tasked with putting together a RFP for a procurement activity. This is where systemigrams come in. When I first heard the word systemigram, I first asked them to say it again, and then spell it, and finally I admitted my ignorance on the topic. To be fair, and in my defense, even for a portmanteau it sounds like it was made up on the spot.
While this may be overwhelming to look at initially, think about how beautifully it represents the multitude of factors that can affect cyber security on social media. This is not a concept that lends itself well to illustration. Systemigrams do an awesome job of organizing chaotic thoughts that relate to complexity.
Blair, Boardman and Sauser wrote about systemigrams as one approach for the United Kingdom’s Ministry of Defense to examine System of Systems (SoS) in lieu of traditional models that were used previously, which typically discounted rather than made sense of the complexity that an organization faces. The process begins with capturing the system in a narrative, and then illustrating it. This concept is hardly new, but the manner in which it is carried out makes it far more useful as it is possible to examine the subject holistically. They listed the following rules when building a systemigram:
Rules for Prose
1. Address strategic intent, not procedural tactics.
2. Be well-crafted, searching the mind of reader and
3. Facilitation and dialogue with stakeholders
(owner/originator of strategic intent) may be required
to create structured text.
4. Length variable but less than 2000 words; scope
of prose must fit scope of resulting systemigram
Rules for Graphic
1. Required entities are nodes, links, inputs, outputs,
2. Sized for a single page.
3. Nodes represent key concepts, noun phrases
specifying people, organizations, groups, artifacts,
4. Links represent relationships and flow between
nodes, verb phrases (occasional prepositional
phrases) indicating transformation, belonging,
5. Nodes may contain other nodes (to indicate
break-out of a document or an organizational/product/process
6. For clarity, the systemigram should contain no
crossover of links.
7. Based on experience, to maintain reasonable
size for presentation purposes, the ratio of
nodes to links should be approximately 1.5.
8. Main flow of systemigram is from top left to
9. Geography of systemigram may be exploited to
elucidate the “why,” “what,” “how” in order to
validate the Transformational aspect of the systemic
10. Color may be used to draw attention to subfamilies
of concepts and transformations.
Even with these rules clearly stated, people struggle with complex concepts. It turns out that the ability to create and interpret systemigrams depend upon abilities that are often called Systems Thinking Skills (STS). Dorani et al. broke these skills down into seven categories: Dynamic Thinking, System-as-Cause Thinking, Forest Thinking, Operational Thinking, Close-Loop Thinking, Quantitative Thinking, and Scientific Thinking. These all boil down to looking at the whole as being more than just the sum of its parts.
This work expanded upon that of Richard Plate who developed CMAST, Cognitive Mapping Assessment of Systems Thinking. Plate used CMAST to evaluate the ability of middle school aged children to understand and interpret non-linear relationships in a complex system. Plate’s work is an awesome read, and one story towards the end really stood out for me. It was about a child that wanted very badly to be correct, but was unable to deviate from a linear structure to a more uncomfortable, but correct, structure that involved branching.
One of the reason that I believe complexity is arguably the single largest threat to any project’s success is the fact that I have met many adults, some in positions of authority, that thought like this child: in a linear manner, unwilling to deviate even though they want to, even though they know that their answer is wrong – they just can’t wrap their minds around the complexity.