2014/01/27 11:00 Dave Alberts, “System Agility IQ”, Agile Systems and Systems Engineering Working Group

Dave Alberts (DoD Command and Control Research Program) with the Agile Systems and SE WG commentary to improve the quality of working paper to be submitted to the July issue of INCOSE Insight

This digest was created in real-time during the meeting, based on the speaker’s presentation(s) and comments from the audience. The content should not be viewed as an official transcript of the meeting, but only as an interpretation by a single individual. Lapses, grammatical errors, and typing mistakes may not have been corrected. Questions about content should be directed to the originator. The digest has been made available for purposes of scholarship, posted by David Ing.

INCOSE International Workshop 2014, Torrance, California


Review articles are at:

Not from the agile community

In the command-and-control community, recognize that overall idea of agile (commanders to maintain order), use systems broadly

System as a collection of interdependent systems.

  • Where do dynamics and complexity come from?
  • The complexity of one in the network can impact the performance of another in the network
  • Wholistic approach can improve performance.
  • Have to give up optimal performance for agility.

Have to give up optimal performance for agility

  • Trying in experiment for some years

Manifest agility

  • Need to be able to measure potential agility.
  • Potential is what IQ is all about
  • A function of enablers of agility
  • Can be done independently of observing a system at a point in time at a particular situation

From a different community, what is meant when someone says agility

Some areas of commonality (three, not sure if more)

  • Agility as a necessary response to complexity and dynamics, otherwise would have a stable situation, or could reduce complexity and optimize
  • Thus, dealing with uncertainties
  • Is it a capability or an outcome?

Is it a capability or an outcome?

  • Was a system agile or not?
  • Could it improves its situation, would say, yes, it’s an agile system; otherwise, say no.
  • It’s not about just the reactive part.
  • Could say have a reactive system, recognize that something will change, will do something to get back on track or could prevent decline.
  • Could also be proactive

Manifest agility:

  • Can observer whether a systems does or does not stay within the acceptable range of performance.
  • Change, stress, when detected, when considered to be signficant, when do you decide to do something, what do you do, when does it take to get back to the desired state?

Depends on coming across situations that require agility, which sometimes happens often or not often.

  • May not have a lot of data
  • Data that you have may not be indicative of data in the future.
  • e.g. currency traders, top picks are all wrong
  • This pattern repeats itself in all domains

Can simulate, but may be hard if there’s a lot of interactions between the system and other systems.

  • Have played without thousands of alternatives.
  • Fidelity issues.
  • Not stressing the scenarios, it’s about what they think might happen.
  • Don’t want to do this blindly, don’t want to do in expectation.
  • Want scenario-independent.

If could only find a way to get agility IQ, then could see if expectations of environments, could see if will deal with unexpected.

We think about agility as an outcome, and the enablers of agility as way to get an outcome we want.

  • If you’re not responsive enough, you’re toast.
  • May be other ways to get you back into the zone of acceptability
  • The white space:  why do we need agility, that pressures us into the zone of acceptability.

Have stressors and enablers.

What do we need to understand agility IQ?

  • Could have more resilience, more responsiveness

For systems engineering community, could try to understand the ways and means by which the mechanism work, to build more flexibility or responsiveness

Recap:

  • Agility as an essential capability
  • Have to be something where you’ll sacrifice optimality
  • Have to observe both enablers and outcomes
  • Measuring real life isn’t good enough, success doesn’t have anything to do with what you did
  • Limitations to scenarios, there’s an additional agility measure

Question: Measures?

Reliability in software and in hardware?  Agility is around change, so reliability may not mean the same.

Question:  Figure 1, why is the lower curve preferred?

Lower curve isn’t preferred.  We can only observe reality.  The lower black curve is reality, the upper red curve is what we would have expected if nothing had changed.

Comment:  Need to change the wording on the chart

Comment:  Passive, reactive and proactive, which is strong.  Later, you don’t use those words, so may fix paragraph.

Question:  Is agile for a system or an evolvement activtity?  System agility is a design for an intended mission.  There’s no agility, e.g. a race car is designed for speed, not for 7 people.  Need to qualify, as system agility versus process agility?

Comment:  Clarify the difference between system agility and process agility.

Comment:  Title is exciting.  However, then think of original IQ to a psychologist as the ability to learn.  Would like to hear more about the learning ability.

Used in a difference sense.  Used it as potential.  Could use a different term?

Innovativeness part as that sense.

Comment:  Change IQ to AQ?

Accepted.

Comment: Clarify what agility means in terms of foreseen change, and unforeseen change, when think in terms of versatility.  A Swiss army knife is versatile, but not agile.

Agree on versatile.  A Swiss army knife is more agile than a screwdriver.

Comment:  Agility onto the poeple who operate the system.  They don’t want autonomous systems.  They want it to be changed to their purposes

When have people adapting to system performance, and then the system adjusting to work practices, there’s an interdependency.

Could need to change organizational parameters to deal with degraded systems.