Every number on this page is computed by the Chapter 2 notebook's own code. Your notes stay on this page until you leave it; copy the memo before you go.
Meridian's CFO has asked for a single definition of “the enterprise” to be used in all board materials on the three options. You are on the finance team. Before recommending a definition, you want to see what each candidate boundary does to the numbers, which parts of the company are coupled to which, and how Meridian's own management rules behave under pressure.
Use AXIOM to test boundary choices on Meridian's map of entities and flows, see how environments change from level to level, and compare a fixed management rule with an adaptive one.
Six entities and nine flows (Table 2.6; Figure 2.2). Industrial Systems sells $2.56 billion a year, 61 percent of Meridian's $4.2 billion, and ships $1.54 billion of it to independent dealers, who resell at about 20 percent more (Worked Example 2.1).
In the boundary designer, place only Meridian inside the perimeter. Record which flows the tool labels inputs, outputs, interior, and environmental.
Drag the dealer network across the boundary and record the new labels. Check the conservation panel: is the total set of flows unchanged? Repeat with another entity of your choice and note which moves change nothing.
In the hierarchy inspector, zoom from group to division to unit. For Digital Solutions, note which parts of Meridian now appear as its environment.
The flows Figure 2.3 draws between divisions, both ways, classified by Definition 2.3 with the boundary drawn at each level.
Digital Solutions' environment is the external environment plus its two sibling divisions, Industrial Systems and Advanced Materials, with their four units (Definition 2.4: a subsystem's environment is its parent's environment plus its siblings). At the group level the coupling between divisions is interior; drawn around one division, it crosses that division's boundary and is environmental for the third.
Definition 2.4; Figure 2.3; notebook cell 11.
On the feedback bench, run the efficiency example with feedback switched off and then on, and vary the feedback gain. Then switch on the adaptation law and shift the environment. Which rule copes better, and which behaves less predictably?
The rule sets each year's effort from the gap: u = gain × (79 − efficiency). Efficiency starts at 71 (Table 3.1). The turnaround's own rule closes 1 − e−0.6 = 0.451 of the gap a year (Chapter 1 model; Worked Example 5.1).
From year 5 the environment takes the shift from efficiency each year. The fixed rule keeps its gain; the adaptive rule raises its gain by the adaptation speed × (squared gap − 1) each year. The shift, speed and tolerance are the notebook's bench settings; the book prints none.
In the loop mapper, build Industrial Systems' migration loop—volume, unit cost, customer migration—and label each link as reinforcing or balancing.
Give each link its sign: does more of the first make more (+) or less (−) of the second?
A loop's polarity is the product of its link signs: an even number of negative links makes it reinforcing, an odd number balancing (Tables 2.4 and 2.5).
Moving the boundary relabels flows but never changes the total (Theorem 2.2); each division's environment includes its siblings (Definition 2.4); feedback holds a target against the environment's pull, while adaptation trades predictability for flexibility (Proposition 2.3).
Recommend a boundary for the board materials, stating which analyses it supports, which it makes harder, and what it cannot affect.