Every number on this page is computed by the Volume II Chapter 12 notebook's own code; the program is deterministic, and the 9,664 payout programs are solved with SciPy's HiGHS, as in the book. Your notes stay on this page until you leave it; copy the memo before you go.
The board will not adopt a single measure for the automation program until it has seen what each owner gives up. The chief financial officer, the chief operating officer, and the treasury have each stated an objective, and the board wants the efficient plans, the prices between them, and a recommendation it can defend. You support the team preparing it.
Use AXIOM to compute the program's frontier, read its prices, bound the reserve, and choose by prices, aspirations, and goals.
Worked Example 12.1: a round in year k costs $120 million and halves the remaining gap to 85 points of efficiency; run-rate savings are $10 million a year per point above 71; payouts are free; cash must stay above $350 million, reach $470 million before a round, and end at $470 million. The CFO judges a plan by the present value of its payouts at 10 percent; the COO by the run-rate savings in year 6; the treasury by the reserve by which every cash commitment is kept.
In the plan explorer, solve the payouts for every schedule of rounds, plot payouts against savings, and mark the dominated plans.
Compute the price of savings between neighbouring efficient plans, and find the range of prices for which each plan wins a weighted sum.
In the bounds panel, raise the reserve, read its multiplier, and find where the schedule of rounds must change.
In the choice panel, set aspirations and weights for a Chebyshev choice, and set goals with and without a second stage.
Each owner's shortfall from the ideal ($384.17 million of payouts, $137.81 million of savings, a $150 million reserve) is measured as a share of the range from the nadir ($22.06 million, $70 million, no reserve); the choice minimizes the largest weighted shortfall plus 10⁻⁴ times their sum, over the 740 efficient plans.
Load Chapter 9's packages and read the price of the board's appetite from the cost–chance frontier.
Chapter 9's cheapest committed packages as Online Appendix Example 10.B.2 prints them (declared values, on Chapter 9's 100,000 paths), and Chapter 11's records' family at 5 percent.
Dominated plans that nobody should propose (Theorem 12.1); prices that decide between neighbouring plans (Theorem 12.2(iv)); a reserve that costs $0.909 a dollar until a round moves; plans in dents that only aspirations reach (Theorem 12.3(iii)); and goals that leave money on the table (Proposition 12.1(ii)).
The efficient plans, the prices between them, the recommended plan with the prices or aspirations that select it, and what each owner gives up.