Stage-3 outputs, and generating real alternatives

Four outputs

Stage 3 — develop and analyse alternatives — is the stage the memo rubric weighs most heavily, because it is where the recommendation is earned. It produces four things: a set of real alternatives; explicit criteria, each traced to a stakeholder objective; an analysis of every alternative against every criterion — in this course, a weighted matrix with a reason in every cell; and a written bias check. The rubric for the stage, R-psp-3 — a Memra rubric that mirrors the stage criteria the open sources agree on — has six lines, and this module takes them in order. The first: at least three genuinely different alternatives, including the status quo where sensible.

Brainstorm before judging

Alternatives are generated in two passes, and the passes must not be mixed. First, generate without judging: list every course of action that would act on the cause found in stage 2, however unlikely. Second, prune: drop what is infeasible against the stage-1 constraints, and merge what is the same action twice. Judging while generating is how a set of alternatives ends up with one real option and two decoys — the reader sees it, and so does the marker.

Worked example — Lakeshore's set

The cause is a failing press. Actions on it:

• A. Keep and repair — the status quo. Repairs were $38,000 last year and rising; nothing else changes. It belongs in the set because it is the baseline every other option is measured against — "better than what?" has to have an answer.
• B. Buy a new press — $260,000 on a bank loan at 7 % over five years, about $63,400 a year, plus about $4,000 a year in repairs: roughly $67,400 a year.
• C. Lease a new press — $58,000 a year for five years, maintenance included.
• D. Outsource large-format work to a Hamilton trade printer — the margin on that $310,000 of work falls from 42 % to 18 %, about $74,400 a year of margin lost, and two press operators become redundant.

A fifth was proposed and rejected: close the large-format line. It is a straw alternative — nobody intends it, it loses on every criterion, and its only function is to make the others look good. Straw alternatives cost a rubric line, because genuinely different means each option could be recommended by someone reasonable.

A sixth was proposed and kept as a candidate: a hybrid — lease a press sized for the base load and outsource the peaks. Hybrids are where the best answer often hides, because they act on two criteria at once (cost and flexibility here). List one whenever the alternatives split along a line that could be combined.

Generate the clinic's set

Cause: a paper book and a phone line. Without judging: keep the paper book (status quo); a reminder-only service by text, book unchanged; a full online-booking system with reminders; a hybrid — online booking for new patients and reminders for all, phone kept for the rest; a second phone line and a third receptionist; move to the dentists' supplier's practice-management suite. Now prune: the third receptionist is infeasible against cost and does not stop double-bookings (still one book) — drop it; the supplier's suite is the full system by another route — merge. Four remain, including the status quo and a hybrid.

Generate cold

> A Kitchener bakery's delivery van, twelve years old, failed twice this month.

Before reading on: keep and repair; replace with a new van (buy); replace with a used van; lease; use a courier for deliveries; a hybrid — courier for the far routes, van for the near. Six, and the status quo is first. An action, not a goal: "improve delivery reliability" is an objective, and it belongs in the criteria, not the alternatives.