2-apphead.gif (6644 bytes)

 

Appendix A: Chaos and Measurement

 

Measurement of Initial Conditions

v_25a.gif (13612 bytes)

In Appendix Figure 1 we are considering that the initial condition of an organization is measured or assessed according to two variables X (e.g., current FTE's) and Y (e.g., number of admissions) graphed in a coordinate system (called a phase or state space). If measurement could have perfect accuracy, then it could be represented in the coordinate system as a "perfect point" (see the point P(x, y) within the smaller shaded square on the left). But because there can never be perfect accuracy in measurement, this "perfect point" would actually have to be a two dimensional region surrounding the "perfect point," i.e., the smaller shaded square on the left of the diagram which represents the inevitable error or lack of knowledge or information we have about the system at t1. The inevitable imprecision inherent in assessment means that there will always exist a certain amount of information that is just not available at any particular time.

Expansion of Measurement Imprecision

v_25b.gif (23478 bytes)

In Appendix Figure 2, we find the smaller shaded square expanding due to sensitive dependence on initial conditions. In a strongly nonlinear system such as chaos, the lack of perfect accuracy in initial measurements becomes increasingly worse, in other words, the information we are missing about the system exponentially increases. - you can see how much larger the shaded box on the right is compared to the smaller one on the left.

Remeasurement at Later Time:

v_26a.gif (20897 bytes)

Consider, for instance, that instead of trying to predict the future based on an assessment or measurement of initial conditions at some point of time, we merely remeasure or reassess a system at a later time (t2) in the same way that we measured or assessed the system at the initial time. We again see the larger shaded square on the right, which originally represented the growth in inaccuracy (or lack of information or knowledge), but, now the remeasurement at time (t2) shrinks the region of imprecision that had been expanded when we, at an earlier time, had projected the initial assessment into the future. The remeasurement shrinks the expanded region because it is an actual measurement, not a projection from a past state which had to expand the region due to sensitive dependence on initial conditions in this strongly nonlinear system. Remeasurement by itself, however, is not sufficient to generate greater information or knowledge of the system; we also need to relate the new to the old measurement; see Appendix Figure 4:

Backward Flow of Decrease of Ignorance:

v_26b.gif (24398 bytes)

Since the remeasurement of the system at the later time shrinks the right hand expanded square to a smaller dimension, there is now greater precision or more information available. That is, our ignorance based on the earlier projection has decreased.

 

Next | Previous
Planners as Nonlinear and Complex Explorers Contents | Applications Contents

 

Copyright © 2001, Plexus Institute Permission
to copy for educational purposes only.