3.11: Milestone 10 – Decision Analysis for Control Selection

Milestone 10 – Decision Modeling: Structuring Risk Tradeoffs Through Tree Analysis

Tool Applied: Decision Tree Analysis
Final Output: Modeled Decision Tree + Mitigation Path Summary Memo

1. Scenario Briefing

MEMO
To: SMDC Risk Strategy & Decision Support Team
From: Kira L. Joshi, Chief Operating Officer, SMDC
Date: Week 11 – Risk Tradeoff Analysis Sprint
Subject: Request for Modeled Decision Tree to Support Control Investment Planning

Team,

As we finalize the risk control plan for the pilot release, we need to look carefully at one of our toughest risk decisions—a situation where the tradeoffs between cost, risk, and speed are real and nontrivial.

We have one control option that is highly effective in theory but expensive or complex to implement. We also have alternative actions that might be “good enough”—or could introduce new risks if misunderstood.

I need a structured decision model to evaluate these options using risk-informed thinking.

Use a decision tree format to map:

• The key decision point
• The available options
• The risks or consequences for each path
• Any known probabilities or assumptions
• Where controls might reduce or shift risk
• Where follow-up decisions may be required

You don’t need to find the “right” answer—you need to make the logic of the decision visible.

We’ll use your tree to support funding allocation and stakeholder alignment. Focus on realism, transparency, and clarity of consequences.

Kira

2. Action Strategy

Purpose of This Milestone

This milestone trains you in risk-based decision modeling using the decision tree format—a tool for:

• Making high-stakes choices visible
• Comparing alternatives across risk and resource dimensions
• Identifying “cascading consequences”
• Supporting collaborative and stakeholder-transparent thinking

In this exercise, you will:

• Select one major control or risk intervention to model
• Identify competing or alternate strategies
• Map out potential outcomes and consequences
• Assign impact or cost values if known
• Summarize findings for team-based planning

This is your first introduction to quantified uncertainty framing.

Step-by-Step Guide

Step 1: Select a High-Stakes Decision Scenario

Choose one risk scenario where:

• Two or more response strategies are being considered
• Tradeoffs between cost, quality, speed, or equity are real
• Stakeholders may be divided on what to do
• You’ve explored this theme in a previous milestone

Examples:

• Alert Overload: Reduce alert frequency (but risk missing true positives) vs. redesign alert language (costly and late)
• API Sync Failures: Build internal redundancy (expensive) vs. wait for vendor fix (low effort but high dependency)
• Data Consent Confusion: Add education module (delays launch) vs. leave consent language unchanged (risk of breach or mistrust)

Clearly label your decision scenario and options.

Step 2: Define the Tree Structure

In a decision tree:

• The root is your decision point
• The branches are your response options
• The nodes are outcomes (success/failure or multiple outcomes)
• Optional: assign probabilities or impact scores to each outcome

Use structured branching like this:

1. Decision: Improve Alert System

• Option A: Redesign language + targeting
  • Outcome 1: Alert clarity improves (80%)
  • Outcome 2: Delayed launch (20%)
• Option B: Reduce frequency only
  • Outcome 1: Alert volume drops, safety risk increases (50%)
  • Outcome 2: No feedback change, clinician trust declines (50%)

Add cost, impact, or likelihood values as known or estimated.

Step 3: Annotate Key Factors

For each branch or outcome:

• Label dependencies (e.g., external vendor support)
• Flag secondary risks (e.g., reputational harm, legal exposure)
• Mark potential decision loops or follow-up actions (e.g., “Revisit in sprint 2 if no adoption”)
• Note equity implications (e.g., “Option A benefits English speakers more”)

Use these annotations to guide your summary memo.

Step 4: Choose Your Decision Recommendation (Optional)

Based on your tree and values:

• Recommend one path, if justified
• OR describe the tradeoffs clearly and neutrally
• OR recommend conditional decision points (e.g., “If usage feedback remains below threshold, escalate to redesign”)

The goal is not to win—it’s to equip the team to choose consciously.

3. Your Deliverable

Part 1: Decision Tree Diagram

• Title and decision scenario stated
• At least two main options (branches)
• At least two outcomes per branch
• Probabilities, impact, or cost labeled if possible
• Dependencies and secondary risks noted
• Final recommendation or “decision path logic” clearly framed

Part 2: Mitigation Path Summary Memo

1–2 page professional memo addressing:

• The decision being modeled
• The control or risk being addressed
• Options and consequences
• Values or assumptions behind the model
• Suggested recommendation or decision path
• Notes on what success or failure would look like

4. Toolkits and Learning Resources

• Decision Tree Modeling Template
• Sample Trees from Health and Tech Projects
• Milestone 9 Control Matrix
• Risk-to-Impact Estimation Guide
• Optional Cost Table or Pilot Budget Framework

5. Critical Reflection

Write 200–300 words addressing:

• What assumptions were hardest to model—and why?
• How did mapping the consequences visually shift your understanding?
• What did you learn about downstream risks or “unspoken costs”?
• What makes decision-making under risk difficult for teams?
• How might this method improve alignment or trust?

6. Quality Control Review

One decision scenario selected and labeled
At least two response paths modeled
Each path includes multiple outcomes
Risks, impacts, or dependencies annotated
Memo is complete, concise, and logically structured
Reflection includes modeling insight and ethical depth
Submission is ready for reuse in Milestone 11 or final integration

7. Final Wrap-Up and Submission

Submit your decision tree, summary memo, and reflection per instructions. This milestone prepares you for Milestone 11, where you’ll address what happens when risks emerge anyway—through contingency planning and escalation design.

You are now structuring the hard decisions—and making risk manageable, even when outcomes are uncertain.