Loading ...
Sorry, an error occurred while loading the content.

10929Re: [scrumdevelopment] RE: Info on Defined and Empirical Processes

Expand Messages
  • Paul Beckford
    Dec 31, 2005
    • 0 Attachment
      Paul Beckford wrote:

      > Mary Poppendieck wrote:
      >
      > > Thanks Hubert.
      > >
      > > The book you mention has its table of contents, first few pages, and
      > index
      > > posted on Amazon.com:
      > >
      > http://www.amazon.com/gp/reader/0195091191/ref=sib_dp_pt/103-7328083-5359068
      > > #reader-link
      > >
      > > Here is what a look there reveals:
      > > "Empirical" is one type of process model, "Defined" does not appear
      > in the
      > > index or the table of contents. The brief section on open loop
      > systems is
      > > in the area of digital control (matches my recollection).
      > >
      > > On page 7, the book says: "Chemical processes are by nature dynamic, by
      > > which we mean that their variables are always changing with time."
      > >
      > > Then the book states that the job of process control is:
      > >
      > > "1. Monitoring certain process indicator variables and
      > > "2. Inducing changes in appropriate process variables in order to
      > improve
      > > process conditions."
      > >
      > > = inspect and adapt.
      > >
      > > In the process industry, what we have called "defined" process control
      > > doesn't really exist; process control is a dynamic response to
      > variation.
      > >
      > > By the way, I was wrong when I gave thermostats as an example of PID
      > > control. The 'gain' in a thermostat is governed by how many degrees
      > > you let
      > > the actual temperature get away from the setpoint before you turn the
      > > heater
      > > or air conditioner on or off, which is pre-set or manually controlled.
      > > However, there is usually no way to vary the amount of heat or cool
      > > that is
      > > produced, so strictly speaking that is not a PID controller. Cruise
      > > control
      > > is a better example of a PID controller.
      > >
      > > Thermostats can be a good example of adaptive control. I have a
      > > thermostat
      > > with a setback - so every night the heater setpoint is automatically set
      > > back several degrees, then it is returned to normal the next morning.
      > > This
      > > thermostat has a program which learns how long it takes to re-heat the
      > > house
      > > in the morning. Over time, the thermostat learns to turn on the furnace
      > > early enough so that it gets up to the desired heat at about the
      > same time
      > > every morning. Changing the parameters of the control function after
      > > it is
      > > first set up is called adaptive control.
      > >
      > > Oh well, control theory is fun for just a very few people, so I will
      > stop
      > > posting this boring stuff. Back to writing about software....
      >
      >
      > No don't stop. If you come from an electronics hardware background like
      > myself it's all very interesting. In electronics they talk about open
      > loop and closed loop systems too just as you describe. If my memory
      > serves me right the term "control" is also reserved for closed loop
      > systems only. The first computing devices, analog operational amplifers
      > where closed loop control systems I believe.
      >
      > I've been threatening to dig up my old electronics stuff since I learnt
      > about closed loop control in software - never got round to it though.
      > Anyone else out there got stuff to add (or corrections, I'm very rusty),
      > I'm more than interested.
      >
      > Paul.

      Something else that came to mind. Modern digital memory is a closed loop
      systems too. At the basis of all digital electronics is a system called
      a bi-stable flip-flop. This is a circuit containing transistors (two I
      think). Now in Mary's description, closed loop control is used to
      maintain an output within a given control limit. To do this the output
      of the system is sensed, a function is applied to it (adapted), then the
      result is negated and summed with the input. This is known as negative
      feedback. With the bi-stable the output is amplified and then summed
      with the input, this is known as positive feedback. The result is that
      the system is only stable in two states, fully on or fully off - hence
      the name bi-stable. This characteristic gives us the 1's and 0's we
      software people rely upon.

      Interestingly the same control principles apply in thermodynamics.
      Basically any system where there are a large number of moving parts
      (elecrons, molecules etc), i.e complexity. That's my lot, other than to
      say that there is a lot of interesting maths associated with this stuff
      that could be applicable to software too.

      A Happy New Year to all .

      Paul.

      >
      > >
      > > Mary Poppendieck
      > > 952-934-7998
      > > www.poppendieck.com
      > >
      > > Date: Sun, 18 Dec 2005 20:33:54 -0700
      > > From: Hubert Smits <hubert.smits@...>
      > > Subject: Re: RE: Info on Defined and Empirical Processes
      > >
      > > Joe, Mary,
      > >
      > > The CSM cours points to the book below when discussing defined vs.
      > > imperical. I haven't read the book, maybe somebody on the list has.
      > >
      > > Process Dynamics, Modeling, and Control
      > > Ogunnaike and Ray, Oxford University Press, 1992
      > >
      > > --Hubert
      > >
      > >
      > >
      > >
      > >
      > > To Post a message, send it to: scrumdevelopment@...
      > > To Unsubscribe, send a blank message to:
      > > scrumdevelopment-unsubscribe@...
      > >
      > >
      > > ------------------------------------------------------------------------
      > > YAHOO! GROUPS LINKS
      > >
      > > * Visit your group "scrumdevelopment
      > > <http://groups.yahoo.com/group/scrumdevelopment>" on the web.
      > >
      > > * To unsubscribe from this group, send an email to:
      > > scrumdevelopment-unsubscribe@yahoogroups.com
      > >
      > <mailto:scrumdevelopment-unsubscribe@yahoogroups.com?subject=Unsubscribe>
      > >
      > > * Your use of Yahoo! Groups is subject to the Yahoo! Terms of
      > > Service <http://docs.yahoo.com/info/terms/>.
      > >
      > >
      > > ------------------------------------------------------------------------
      > >
      >
      >
      >
      >
      > To Post a message, send it to: scrumdevelopment@...
      > To Unsubscribe, send a blank message to:
      > scrumdevelopment-unsubscribe@...
      >
      >
      >
      > SPONSORED LINKS
      > Scrum
      > <http://groups.yahoo.com/gads?t=ms&k=Scrum&w1=Scrum&c=1&s=11&.sig=KvDTKhw7ncC9XbB25jdApQ>
      >
      >
      >
      > ------------------------------------------------------------------------
      > YAHOO! GROUPS LINKS
      >
      > * Visit your group "scrumdevelopment
      > <http://groups.yahoo.com/group/scrumdevelopment>" on the web.
      >
      > * To unsubscribe from this group, send an email to:
      > scrumdevelopment-unsubscribe@yahoogroups.com
      > <mailto:scrumdevelopment-unsubscribe@yahoogroups.com?subject=Unsubscribe>
      >
      > * Your use of Yahoo! Groups is subject to the Yahoo! Terms of
      > Service <http://docs.yahoo.com/info/terms/>.
      >
      >
      > ------------------------------------------------------------------------
      >
    • Show all 16 messages in this topic