I’ve been trying for quite a while now to come up with a diagram that shows how we control variables that we can’t directly affect. I think I’ve got it. Comments anyone?
Regards,
Fred Nickols
Managing Partner
Distance Consulting LLC
“Assistance at A Distance”
[From: Richard Pfau (2019.01.15 17:57 EST)]
Fred,
My first impression is that your diagram seems OK, but it indicates how difficult it can be to achieve targets so indirectly since so many other variables/disturbances may also affect what we are trying to influence or affect. I personally prefer to think that the diagram indicates how we may try to influence variables that we can’t directly affect, rather than indicate as you do that the diagram shows how we control such variables. When so many other disturbances are also occurring, we aren’t really controlling the target are we? We are only attempting to influence it.
Rich
On Tue, Jan 15, 2019 at 2:36 PM Fred Nickols csgnet@lists.illinois.edu wrote:
I’ve been trying for quite a while now to come up with a diagram that shows how we control variables that we can’t directly affect. I think I’ve got it. Comments anyone?
Regards,
Fred Nickols
Managing Partner
Distance Consulting LLC
“Assistance at A Distance”
[Martin Taylor 2019.01.15.23.01]
I'm afraid I don't really understand this diagram. On the one hand,
if you take the figure of a person a the comparator of a control
loop, as the two inputs “Intended Value” and “Perceived Value” seem
to indicate, you have simply drawn a picture of four levels of the
control hierarchy without any many-to-many linkages. On the other
hand, if all these units are in the environment and the word “direct
actions” is at the boundary of the environment (i.e. simply muscular
actions) then you have drawn a set of tools, such as yellow->put
your foot on a pedal, grey-green->increase and decrease the revs
of a motor, blue->move the car at variable speeds, with the
target being your perception that you are where you want to go. But
this doesn’t seem to represent a problem, so I guess I don’t know
what problem you are addressing.
Maybe a real-life example that isn’t analogous to my example might
help me to understand the issue.
Martin
On 2019/01/15 2:34 PM, Fred Nickols
( via csgnet Mailing List) wrote:
I've been trying for quite a while now to come upwith a diagram that shows how we control variables that we can’t
directly affect. I think I’ve got it. Comments anyone?Regards,
Fred Nickols
Managing Partner
Distance Consulting LLC
“Assistance at A Distance”
[Martin Taylor 2019.01.15.23.01]
I'm afraid I don't really understand this diagram. On the one hand,
if you take the figure of a person as the comparator of a control
loop, as the two inputs “Intended Value” and “Perceived Value” seem
to indicate, you have simply drawn a picture of four levels of the
control hierarchy without any many-to-many linkages. On the other
hand, if all these units are in the environment and the word “direct
actions” is at the boundary of the environment (i.e. simply muscular
actions) then you have drawn a set of tools, such as yellow->put
your foot on a pedal, grey-green->increase and decrease the revs
of a motor, blue->move the car at variable speeds, with the
target being your perception that you are where you want to go. But
this doesn’t seem to represent a problem, so I guess I don’t know
what problem you are addressing.
Maybe a real-life example that isn’t analogous to my example might
help me to understand the issue.
Martin
On 2019/01/15 2:34 PM, Fred Nickols
( via csgnet Mailing List) wrote:
I've been trying for quite a while now to come upwith a diagram that shows how we control variables that we can’t
directly affect. I think I’ve got it. Comments anyone?Regards,
Fred Nickols
Managing Partner
Distance Consulting LLC
“Assistance at A Distance”
Rich:
I think in many cases the best we can do is influence those variables we wish to affect. In some cases, however, our influence is sufficient to get what we want. In any case, it is indeed difficult much of the time. Thanks for the feedback.
Regards,
Fred Nickols
Managing Partner
Distance Consulting LLC
“Assistance at A Distance”
Thanks, Martin. I was trying to show the path that leads from our actions to a variable that we cannot directly affect but wish to control or at least significantly influence. In PCT parlance I think that path is known as “the feedback function.” I call it the Achievement Path or Solution Path. Again, in PCT parlance, the feedback function can be simple and straightforward or quite complex, depending on the variable you are trying to influence/control. In any case, I’ll keep working on it. Thanks again.
Regards,
Fred Nickols
Managing Partner
Distance Consulting LLC
“Assistance at A Distance”
Fred Nickols (2019.01.16.0834 ET)
In my scheme of things I refer to what BIll Powers called “the feedback function” as The Achievement Path, the path that runs from actions to the target variable. A couple of Knowledge Worker columns might help clarify things. One sets for the basic concept and the other lays out a workplace example. Here are the links:
https://www.nickols.us/The_Achievement_Path.pdf
https://www.nickols.us/AchievementPathExample.pdf
In the diagram I sent out I used Accessible, Connecting and Driver as the three types of variables between actions and the target variable. In the two columns I used Proximate, Intermediate and Penultimate. They are the same things.
Regards,
Fred Nickols
Managing Partner
Distance Consulting LLC
“Assistance at A Distance”
Fred,
My oppinon is that you upgraded your diagram from “controlled variable” in environment to many of them. Explain to me how it fits into diagram LCS III, beside “PCT Parlance” ? Why inventing hot water ?
Best regards,
Boris
From: Fred Nickols (fwnickols@gmail.com via csgnet Mailing List) csgnet@lists.illinois.edu
Sent: Tuesday, January 15, 2019 8:34 PM
To: csgnet csgnet@lists.illinois.edu
Subject: Living Control System Diagram
I’ve been trying for quite a while now to come up with a diagram that shows how we control variables that we can’t directly affect. I think I’ve got it. Comments anyone?
Regards,
Fred Nickols
Managing Partner
Distance Consulting LLC
“Assistance at A Distance”
[Rick Marken 2019-01-16_17:09:45]
FN: I’ve been trying for quite a while now to come up with a diagram that shows how we control variables that we can’t directly affect. I think I’ve got it. Comments anyone?
RM: Hi Fred. Here’s some questions before I make any comments:
BestÂ
Rick
On Tue, Jan 15, 2019 at 11:36 AM Fred Nickols csgnet@lists.illinois.edu wrote:
–
Richard S. MarkenÂ
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery
Fred Nickols (2019.01.17.0607 ET)
 It’s pretty much what you describe, Rick. Thanks for the pointer to Bill’s demo.  “Remote effect� sounds right. I will check it out.
Fred Nickols
Distance Consulting LLC
“Assistance at A Distance�
www.nickols.us
Fred Nickols (2019.01.17.0725 ET)
Some answers to your questions, Rick
For a concrete example, see www.nickols.us/The_Achievement_Path.pdf
The lines depict different disturbances affecting different variables along the way.
The disturbances do indeed vary slowly.Â
By the way, the demo was fascinating. I could see tha each pulley presented a disturbance and so did their combined effects. Is that right?
You have your third question right.Â
Fred Nickols
Distance Consulting LLC
“Assistance at A Distance�
www.nickols.us
[Rick Marken 2019-01-17_12:08:53]
Fred Nickols (2019.01.17.0725 ET)
FN: Some answers to your questions, Rick
FN: For a concrete example, see www.nickols.us/The_Achievement_Path.pdf
RM: I think this example – of changing the setting of the thermostat to control for the “target variable”, room temperature – is not quite matched by your diagram. The main difference is that the intermediate variables in this “achievement path” have lines looping back to the perceptual value. And these inputs are shown to mix in with the perception of the target variable. If you eliminate these lines then you have a correct picture of what is going on. The components of the “achievement path” are completely invisible to the controller. These components are the electrical circuitry in the thermostat itself and the heater/A/C that produces the output that affect the target variable (air temperature). There are disturbances that can affect these components but steps must be taken to prevent such potential disturbances from affecting these variables. If the disturbances do have an effect on the intermediate variables, you will know this only form the fact that the thermostat doesn’t work. In other works, what you are describing as an “achievement path” is just the feedback connection between the controller’s output and the state of the controlled variable. And all we typically know about the feedback connection between our outputs and the variables we control is that they exist. And we know they exist if we can see that we are able to have systematic effects on a variable we want to control.
RM: I think what you may have in mind as the “achievement path”, given the way you talk about it in the paper, is sequence or program perception that, when controlled successfully, has the the desired ending.Â
FN: The lines depict different disturbances affecting different variables along the way.
RM: The lines going from disturbances to intermediate variables is fine; it’s the line going from the intermediate variables to perceptual input that make no sense.Â
Â
FN: By the way, the demo was fascinating. I could see that each pulley presented a disturbance and so did their combined effects. Is that right?
RM: Yes, the varying positions of the two pulleys are a disturbance to the position of the top end of the line – the controlled variable. These disturbances have simultaneous effects on the “remote” controlled variable. The net effect of these disturbances is compensated for by simply varying the position of the bottom end of the line. You could trace out an “achievement path” from output to controlled variable via the pulleys. But this path is irrelevant to the controller in the sense that he needs to know nothing about it. The controller simply varies output as necessary to protect the controlled variable from the net effect of disturbances, whatever their source.
BestÂ
Rick
You have your third question right.Â
On Thu, Jan 17, 2019 at 6:08 AM Fred Nickols fwnickols@gmail.com wrote:
Fred Nickols (2019.01.17.0607 ET)
 It’s pretty much what you describe, Rick. Thanks for the pointer to Bill’s demo.  “Remote effect� sounds right. I will check it out.
On Wed, Jan 16, 2019 at 8:18 PM Richard Marken csgnet@lists.illinois.edu wrote:
[Rick Marken 2019-01-16_17:09:45]
On Tue, Jan 15, 2019 at 11:36 AM Fred Nickols csgnet@lists.illinois.edu wrote:
FN: I’ve been trying for quite a while now to come up with a diagram that shows how we control variables that we can’t directly affect. I think I’ve got it. Comments anyone?
RM: Hi Fred. Here’s some questions before I make any comments:
- Could you give a concrete example of what your diagram represents?Â
- Assuming that the “Target Variable” is the controlled variable, why are there lines going from each disturbance into the perceptual function?Â
- It looks like the little circles are variables affected by different disturbances. Is that right?
- If the disturbances vary independently, the “Direct Actions”, which I presume are outputs of the controller, will have no regular effect on the “Target Variable”. The only way of dealing with this, I think, would be for the controller to be using an e. coli type control process. And even this process is unlikely to produce any kinds of good control unless the disturbances vary very slowly. Is this what you have in mind?Â
RM: My main question is really the first one: what real-world behavioral situation is this diagram designed to represent. If it’s one where the system output goes through a sequence of causal connection to produce a determinate effect on the controlled variable, then the ability to control in this situation is nicely demonstrated in one of Powers’ demos (http://www.pct-labs.com/tutorial1/index.html); see Step K: Control of Remote Effect. Â
BestÂ
Rick
–
Richard S. MarkenÂ
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery
–
Fred Nickols
Distance Consulting LLC
“Assistance at A Distance�
www.nickols.us
–
Fred Nickols
Distance Consulting LLC
“Assistance at A Distance�
www.nickols.us
–
Richard S. MarkenÂ
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery
Fred Nickols (2019.01.17.1522 ET)
Well, if you can find time, take a look at the attached and let me know if the example better matches the diagram.
Regards,
Fred Nickols
Managing Partner
Distance Consulting LLC
“Assistance at A Distance”
A Conversation about Living Control Systems.docx (72.5 KB)
On Thu, Jan 17, 2019 at 3:13 PM Richard Marken csgnet@lists.illinois.edu wrote:
[Rick Marken 2019-01-17_12:08:53]
Fred Nickols (2019.01.17.0725 ET)
FN: Some answers to your questions, Rick
FN: For a concrete example, see www.nickols.us/The_Achievement_Path.pdf
RM: I think this example – of changing the setting of the thermostat to control for the “target variable”, room temperature – is not quite matched by your diagram. The main difference is that the intermediate variables in this “achievement path” have lines looping back to the perceptual value. And these inputs are shown to mix in with the perception of the target variable. If you eliminate these lines then you have a correct picture of what is going on. The components of the “achievement path” are completely invisible to the controller. These components are the electrical circuitry in the thermostat itself and the heater/A/C that produces the output that affect the target variable (air temperature). There are disturbances that can affect these components but steps must be taken to prevent such potential disturbances from affecting these variables. If the disturbances do have an effect on the intermediate variables, you will know this only form the fact that the thermostat doesn’t work. In other works, what you are describing as an “achievement path” is just the feedback connection between the controller’s output and the state of the controlled variable. And all we typically know about the feedback connection between our outputs and the variables we control is that they exist. And we know they exist if we can see that we are able to have systematic effects on a variable we want to control.
RM: I think what you may have in mind as the “achievement path”, given the way you talk about it in the paper, is sequence or program perception that, when controlled successfully, has the the desired ending.Â
FN: The lines depict different disturbances affecting different variables along the way.
RM: The lines going from disturbances to intermediate variables is fine; it’s the line going from the intermediate variables to perceptual input that make no sense.Â
Â
FN: By the way, the demo was fascinating. I could see that each pulley presented a disturbance and so did their combined effects. Is that right?
RM: Yes, the varying positions of the two pulleys are a disturbance to the position of the top end of the line – the controlled variable. These disturbances have simultaneous effects on the “remote” controlled variable. The net effect of these disturbances is compensated for by simply varying the position of the bottom end of the line. You could trace out an “achievement path” from output to controlled variable via the pulleys. But this path is irrelevant to the controller in the sense that he needs to know nothing about it. The controller simply varies output as necessary to protect the controlled variable from the net effect of disturbances, whatever their source.
BestÂ
Rick
You have your third question right.Â
On Thu, Jan 17, 2019 at 6:08 AM Fred Nickols fwnickols@gmail.com wrote:
Fred Nickols (2019.01.17.0607 ET)
 It’s pretty much what you describe, Rick. Thanks for the pointer to Bill’s demo.  “Remote effect� sounds right. I will check it out.
On Wed, Jan 16, 2019 at 8:18 PM Richard Marken csgnet@lists.illinois.edu wrote:
[Rick Marken 2019-01-16_17:09:45]
On Tue, Jan 15, 2019 at 11:36 AM Fred Nickols csgnet@lists.illinois.edu wrote:
FN: I’ve been trying for quite a while now to come up with a diagram that shows how we control variables that we can’t directly affect. I think I’ve got it. Comments anyone?
RM: Hi Fred. Here’s some questions before I make any comments:
- Could you give a concrete example of what your diagram represents?Â
- Assuming that the “Target Variable” is the controlled variable, why are there lines going from each disturbance into the perceptual function?Â
- It looks like the little circles are variables affected by different disturbances. Is that right?
- If the disturbances vary independently, the “Direct Actions”, which I presume are outputs of the controller, will have no regular effect on the “Target Variable”. The only way of dealing with this, I think, would be for the controller to be using an e. coli type control process. And even this process is unlikely to produce any kinds of good control unless the disturbances vary very slowly. Is this what you have in mind?Â
RM: My main question is really the first one: what real-world behavioral situation is this diagram designed to represent. If it’s one where the system output goes through a sequence of causal connection to produce a determinate effect on the controlled variable, then the ability to control in this situation is nicely demonstrated in one of Powers’ demos (http://www.pct-labs.com/tutorial1/index.html); see Step K: Control of Remote Effect. Â
BestÂ
Rick
–
Richard S. MarkenÂ
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery
–
Fred Nickols
Distance Consulting LLC
“Assistance at A Distance�
www.nickols.us
–
Fred Nickols
Distance Consulting LLC
“Assistance at A Distance�
www.nickols.us
–
Richard S. MarkenÂ
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.�
                --Antoine de Saint-Exupery
[Martin Taylor 2019.01.17.15.24]
Rick, doesn't this last paragraph, which seems correct to me,
contradict what you said earlier: " " I seem to remember somewhere that you or Bill, or maybe someone
else, tried varying the loop gain by putting a disturbance into the
pat between joystick and cursor in a tracking study. Or maybe I’m
imagining it. But having disturbances along Fred’s “Achievement
path” is not likely to eliminate control until, like a direct
disturbance, they happen too fast or too strongly to be countered.
If, say, the power supply to the house is reduced in voltage
erratically, that’s not going to prevent the thermostat from sending
a “more heat” signal to the furnace. It will just change the rate at
which the furnace produces heat (and not even that if it is a gas
furnace rather than an electrical one).
Martin
[Rick Marken 2019-01-17_12:08:53]
Fred Nickols (2019.01.17.0725 ET)
FN: Some answers to your questions, Rick
FN: For a concrete example, see www.nickols.us/The_Achievement_Path.pdf
RM: I think this example -- of changing the setting ofthe thermostat to control for the “target variable”, room
temperature – is not quite matched by your diagram. The
main difference is that the intermediate variables in this
“achievement path” have lines looping back to the perceptual
value. And these inputs are shown to mix in with the
perception of the target variable. If you eliminate these
lines then you have a correct picture of what is going on.
The components of the “achievement path” are completely
invisible to the controller. These components are the
electrical circuitry in the thermostat itself and the
heater/A/C that produces the output that affect the target
variable (air temperature). There are disturbances that can
affect these components but steps must be taken to prevent
such potential disturbances from affecting these variables.
If the disturbances do have an effect on the intermediate
variables, you will know this only form the fact that the
thermostat doesn’t work. In other works, what you are
describing as an “achievement path” is just the feedback
connection between the controller’s output and the state of
the controlled variable. And all we typically know about the
feedback connection between our outputs and the variables we
control is that they exist. And we know they exist if we can
see that we are able to have systematic effects on a
variable we want to control.
RM: I think what you may have in mind as the "achievementpath", given the way you talk about it in the paper, is
sequence or program perception that, when controlled
successfully, has the the desired ending.
FN: The lines depict different disturbancesaffecting different variables along the way.
RM: The lines going from disturbances to intermediatevariables is fine; it’s the line going from the intermediate
variables to perceptual input that make no sense.
FN: By the way, the demo was fascinating. Icould see that each pulley presented a disturbance and so
did their combined effects. Is that right?
RM: Yes, the varying positions of the two pulleys are adisturbance to the position of the top end of the line –
the controlled variable. These disturbances have
simultaneous effects on the “remote” controlled variable.
The net effect of these disturbances is compensated for by
simply varying the position of the bottom end of the line.
You could trace out an “achievement path” from output to
controlled variable via the pulleys. But this path is
irrelevant to the controller in the sense that he needs to
know nothing about it. The controller simply varies output
as necessary to protect the controlled variable from the net
effect of disturbances, whatever their source.
These components are the
electrical circuitry in the thermostat itself and the heater/A/C
that produces the output that affect the target variable (air
temperature). There are disturbances that can affect these
components but steps must be taken to prevent such potential
disturbances from affecting these variables. If the disturbances
do have an effect on the intermediate variables, you will know
this only form the fact that the thermostat doesn’t work.*
[Martin Taylor 2019.01.17.`5.42]
Fred Nickols (2019.01.17.1522 ET)
Well, if you can find time, take a look at the attached and let me know if the example better matches the diagram.
If I may make a suggestion, you could use the coffee example all the way through, rather than the complex management filing problem. You can't fill the coffee cup because there is no coffee ready. You don't know how the coffee maker works (or coffee is supplied by the canteen on request), so you ask someone to make it (or supply it). Disturbances to that process of getting coffee supplied ready to pour might include the provider getting involved in unrelated conversation, or coffee grounds having run out and that person asking an errand boy to go to the store-room and get some, etc. etc. In the end, your achievement path includes coffee ready to pour, and the error in the target variable is corrected when the cup is full.
Does this describe what you want to get across? I got the gist, but not the details of your "load rate" example. Maybe your intended audience are more familiar with that environment and will now immediately what you are talking about.
Separate comment: In page 3 you mention the "feedback loop" closing the loop, which I think could be made clearer by calling it a "feedback path" or a "feedback connection" rather than "feedback loop".
Martin
Fred Nickols (2019.01.17.1601 ET)
Thanks, Martin  it might make sense to use the coffee example. That would also shorten the piece.
Fred Nickols
Distance Consulting LLC
âAssistance at A Distanceâ?
www.nickols.us
[Rick Marken 2019-01-17_15:53:50]
[Martin Taylor 2019.01.17.15.24]
RM: Yes, the varying positions of the two pulleys are adisturbance to the position of the top end of the line –
the controlled variable. These disturbances have
simultaneous effects on the “remote” controlled variable.
The net effect of these disturbances is compensated for by
simply varying the position of the bottom end of the line.
You could trace out an “achievement path” from output to
controlled variable via the pulleys. But this path is
irrelevant to the controller in the sense that he needs to
know nothing about it. The controller simply varies output
as necessary to protect the controlled variable from the net
effect of disturbances, whatever their source.
MT: Rick, doesn't this last paragraph, which seems correct to me,contradict what you said earlier: "* These components are the
electrical circuitry in the thermostat itself and the heater/A/C
that produces the output that affect the target variable (air
temperature). There are disturbances that can affect these
components but steps must be taken to prevent such potential
disturbances from affecting these variables. If the disturbances
do have an effect on the intermediate variables, you will know
this only form the fact that the thermostat doesn’t work.* "MT: I seem to remember somewhere that you or Bill, or maybe someoneelse, tried varying the loop gain by putting a disturbance into the
pat between joystick and cursor in a tracking study. Or maybe I’m
imagining it. But having disturbances along Fred’s “Achievement
path” is not likely to eliminate control until, like a direct
disturbance, they happen too fast or too strongly to be countered.
If, say, the power supply to the house is reduced in voltage
erratically, that’s not going to prevent the thermostat from sending
a “more heat” signal to the furnace. It will just change the rate at
which the furnace produces heat (and not even that if it is a gas
furnace rather than an electrical one).
RM: I did set up a control simulation where the gain of the feedback function varied. That is, qi = k*o where k was a variable. This had no detrimental effect on control because it preserves the monotonic relationship between input and output, same as in Bill’s Remote Effects demo where the changes in the positions of the pulleys just change the gain of the feedback relationship between output and input. But I did a simulation of Fred’s “achievement path” model and I was unable to get control of the target variable. I believe it’s because there is not a monotonic relationship between output and input in this situation.Â
RM:Â I think this “achievement path” idea is not consistent with a PCT understanding of how control works. But I’ve got other things to do for a few weeks so I’ll try to avoid net discussions and you can so carry on as you wish. I’ll just suggest taking seriously the idea of control of a hierarchy of perceptions, where higher level perceptions can be things like sequences and programs. I believe the “achievement path” idea could be consistent with PCT if it were understood (and diagrammed) as a sequence or program perception, controlled by means of varying lower level perceptions.
Best
Rick
Â
Richard S. MarkenÂ
"Perfection is achieved not when you have nothing more to add, but when you
have nothing left to take away.â?
                --Antoine de Saint-Exupery
[Eetu Pikkarainen 2019-01-18_10:42:25 UTC]
[Martin Taylor 2019.01.17.`5.42]
Fred Nickols (2019.01.17.1522 ET)
Fred, that is a nice dialogue!
I much agree with Martin's comments. I would still add that in (at least some) PCT diagrams the "feedback function" is situated between the output of the controller and the input quantity (which is some times called "controlled variable" or in Fred's case "target variable"). This is one peculiarity of PCT view: the subject gives feedback to the environment (about how it should be). In humanistic-cognitive tradition the subject gets feedback from the environment about how the plans of the subject realize.
So for me it reminds more about TOTE type of models than PCT type to call the line from target variable to subject "feedback path". I would rather call it e.g. "information path".
However, if we stress the idea of closed loop and control of perception, then we should say that the subjects gives feedback to its own perception how it should be. According to this idea the whole chain of effects from the output function to the input function is the feedback path. It has two parts: first the output path ("solution path") i.e. chain of effects from the output function to the target variable (or CEV) and second the input path ("information path") i.e. the chain of effects from the target variable to the input function. BOTH of these paths can be either simple or complex. To get the perception of the value of the target variable you may have to use many kind of methods, instruments, surveys, snoopers, spies, reports etc.
Rick in [Rick Marken 2019-01-17_12:08:53] noted that "it's the line going from the intermediate variables to perceptual input that make no sense". I somewhat understand his point but I think that your diagram is a combination of more than one control loops which are nested. You control for the target variable but to do that you have to first control for an accessible variable. You have a perception of the type of the filing system and you have a reference for a different perception. Those connecting variables are not accessible for you to affect them but they can be accessible for you to perceive them. If so then you probably have references for them, too, and then the line from them to you perception makes sense. Some of them might be hard or impossible to perceive (otherwise than via the changes in the target variable) and then the line does not make sense.
Eetu
Well, if you can find time, take a look at the attached and let me
know if the example better matches the diagram.
If I may make a suggestion, you could use the coffee example all the way through, rather than the complex management filing problem. You can't fill the coffee cup because there is no coffee ready. You don't know how the coffee maker works (or coffee is supplied by the canteen on request), so you ask someone to make it (or supply it). Disturbances to that process of getting coffee supplied ready to pour might include the provider getting involved in unrelated conversation, or coffee grounds having run out and that person asking an errand boy to go to the store-room and get some, etc. etc. In the end, your achievement path includes coffee ready to pour, and the error in the target variable is corrected when the cup is full.
Does this describe what you want to get across? I got the gist, but not the details of your "load rate" example. Maybe your intended audience are more familiar with that environment and will now immediately what you are talking about.
Separate comment: In page 3 you mention the "feedback loop" closing the loop, which I think could be made clearer by calling it a "feedback path" or a "feedback connection" rather than "feedback loop".
Martin
Thanks, Eetu. I’ll keep working on it.
Regards,
Fred Nickols
Managing Partner
Distance Consulting LLC
“Assistance at A Distance”