(Gavin Ritz 2008.04.20.10.17NZT)
Arthur
PCT and VSM are not compatible (they have very different premises) and PCT
in itself has no applicable model to do what you ask. VSM has also no
appropriate methodology to do this even Beer resigned himself to this. His
masterpiece for practical application is Diagnosing the System, and
Designing Freedom, not very good at either diagnosing or designing.
Control in organisation is vested in the roles and it's the inter roles that
are required to be so called "authori-tised" with the appropriate
authorities assigned to that role, specifically the cross functional roles.
I suggest you look at Elliot Jaques Requisite Organization for a glimpse
into seeing how to do this. Buy Jaques books on Executive Leadership and his
book on Requisite Organization and it will tell exactly how to do what you
want to achieve.
Ultimately any complexity type approach forces you to decide what is the
"parts and what is the whole". Parts and wholes are often different
depending on what you are trying to achieve.
Controlling at an organisational level really just involves- "stopping",
getting someone to "do something" with parameters, delaying, deciding,
reporting, persuading, informing and not much else. Take these few
authorities of control and ask the question of your own role and the role of
another, see what you think your authorities are in terms of "your role vs.
her/his role) and then ask the other person the same thing.
You will be shocked to know that they will differ hugely. If you have not
defined the controlling parameters I mentioned above. (Is that not the
source of a massive conflict)? Ask this question, your colleague at the same
of level organisation as you in marketing what does he think he can tell you
to do and what do you think you can tell him and he is compelled to do it?
Almost 99% of organisations have no idea how controlling works and don't
have any such authorities in place to manage controlling. In fact it's the
greatest source of conflict in organisations. Most firms will have financial
authorities and no people authorities.
Gavin
Dear Bill,
Thanks for your response. I agree that the operationalisation of all these
concepts is a critical and difficult task. I just bought all your books and
will explore how PCT and VSM can maybe be supplemental.
To what extend can the PCT concept be transferred to an organisational
environment like an airline ? Has the PCT concept explicitly been used to
model the control structure of a company ? It would be helpful to read about
this.
Thanks for any suggestions or links.
Arthur
[From Bill Powers (2009.04.19.0-712 MDT)]
AD:
I am trying to use these concepts including the Viable System Model
for the description of a safety management system so your (and
others of course) comments are very welcome.
BP:
Tell me something. Suppose you design a system for managing safety,
and it doesn't work as well as you want. You hire a consultant, and
he reports to you that your system's output actions don't have
enough variety to match the variety of the environment you're trying
to control. What does that tell you about how to make the system work
better?
AD:
That would be a very underspecified report and no help at all. I
would ask him on what specific dimensions the system lacks variety.
Then I could think of measures to amplify that specific system
variety, or how to attenuate the environmental variety impacting the
system, to recreate homeostasis.
BP: Fine, but you need to know something about control systems to do
all that successfully. Behind what you call "amplifying variety" is
something much more specific: identifying all the variables that need
to be controlled, and providing some means of affecting each of those
variables. That's all that "requisite variety" means. And you need to
develop some way to monitor the results to see if you're getting the
result you want or something else. That's the rest of the control
system. In fact, to find those variables and figure out how to
control them doesn't require thinking about variety at all, though
there's nothing to stop you from doing so if you wish, after you've
solved all the real problems. As you say, variety is a rather
underspecified concept.
Just trying to eliminate disturbances won't "recreate homeostasis,"
either. You can't eliminate all disturbances, especially in
airplanes. What you need is an actual homeostatic system or as we
call them here, a control system. And it had better not be just a
homeostatic system; what you need is a RHEOstatic system (as
Mrosovski calls them), better known as a hierachy of control systems
in PCT circles. You wouldn't want an autopilot that could only keep
you, homeostatically, at one heading, speed, and altitude. First you
need to be able to maintain each important variable in a specific
state, and then you need to be able to vary the state in which each
variable is being controlled, so the lower-order control systems can
be put into use by more general, higher-order control systems. The
means of varying the homeostatic state is what we call a reference
signal, and that is how higher-order systems can change what lower
systems are doing without coming into conflict with them. The higher
systems tell the lower ones what state to maintain, and leave the
actual maintaining up to them. It's like the pilot entering the
desired heading, airspeed, and altitude into the autopilot. He
doesn't tell the autopilot how to manipulate the ailerons, elevators,
and throttle -- he just tells it what result to achieve. If he tried
to operate the controls he would be fighting the autopilot. There's a
reason for making the autopilot cut out if the pilot starts using the
controls himself. In living systems, that sort of micromanagement
isn't allowed.
Every recursion of the system (VSM language) must have req var and
this can be achieved by variety amplification and attenuation. I
guess you are probably more familiar with these concepts than I am.
No, actually I don't use those terms at all. What I do in designing
control systems or models of them could probably be classified in
such abstract terms, but it's not the abstract terms that do the
heavy lifting. I design control systems and make them work, and never
once even think about variety. So far that hasn't proved to be a
serious omission.
I am an airline pilot and the idea of a a crew matching the
environmental variety to maintain essential variables (e.g. speed,
altitude, direction) within limits feels as a useful model.
Perhaps that works for you, but I also recommend seeing how PCT looks
to you as a model of those processes. You're making me wonder if we
don't need yet another book, something like "How to use PCT in the real
world."
Best,
Bill P.
···
At 10:40 PM 4/18/2009 +0200, Arthur Dykstra wrote: