6.6 Problem Solving and Decision Making

6.6 Problem Solving and Decision Making

1. Problem-Solving Process

Problem Identification Phase

1. Problem Recognition:

   • Detecting deviations from expected performance.

   • Identifying symptoms vs root causes.

   • Gathering initial data and observations.

2. Problem Definition:

   • Clearly stating the problem in specific terms.

   • Establishing problem boundaries and constraints.

   • Determining what constitutes a "solution."

3. Goal Setting:

   • Defining success criteria.

   • Establishing measurable objectives.

   • Setting realistic timelines.

Analysis and Solution Development

1. Root Cause Analysis:

   • 5 Whys Technique: Asking "why" repeatedly to drill down.

   • Fishbone Diagram (Ishikawa): Categorizing potential causes (Man, Machine, Method, Material, Measurement, Environment).

   • Fault Tree Analysis: Logical diagram of failure pathways.

2. Information Gathering:

   • Collecting relevant data.

   • Researching similar problems.

   • Consulting experts and stakeholders.

3. Solution Generation:

   • Developing multiple potential solutions.

   • Considering various approaches.

   • Evaluating feasibility of each option.

Implementation and Evaluation

1. Solution Selection:

   • Applying decision-making tools.

   • Considering risk, cost, and benefits.

   • Choosing the optimal solution.

2. Implementation Planning:

   • Developing action steps.

   • Assigning responsibilities.

   • Creating timelines and milestones.

3. Execution:

   • Implementing the chosen solution.

   • Monitoring progress.

   • Making adjustments as needed.

4. Evaluation and Learning:

   • Measuring outcomes against goals.

   • Documenting results and lessons learned.

   • Standardizing successful solutions.

2. Brainstorming

Preparation Phase

1. Define Clear Objective:

   • Specific problem statement.

   • Desired outcomes.

   • Constraints and limitations.

2. Assemble Team:

   • Diverse perspectives (5-10 people optimal).

   • Mix of expertise levels.

   • Include both experts and fresh thinkers.

3. Set Ground Rules:

   • No criticism during idea generation.

   • Encourage wild ideas.

   • Build on others' ideas.

   • Aim for quantity over quality initially.

Idea Generation Techniques

1. Traditional Brainstorming:

   • Free-flowing verbal idea sharing.

   • Round-robin or open format.

   • Time-limited sessions (15-45 minutes).

2. Brainwriting (6-3-5 Method):

   • 6 people write 3 ideas each in 5 minutes.

   • Pass papers and build on others' ideas.

   • Silent, reduces dominance by vocal participants.

3. Structured Approaches:

   • SCAMPER Technique:

     • Substitute

     • Combine

     • Adapt

     • Modify

     • Put to another use

     • Eliminate

     • Reverse

   • Morphological Analysis: Breaking problem into parameters and exploring combinations.

Post-Brainstorming Activities

1. Idea Capture:

   • Record all ideas without filtering.

   • Use whiteboards, sticky notes, or digital tools.

   • Assign unique identifiers to each idea.

2. Initial Filtering:

   • Group similar ideas (affinity diagramming).

   • Eliminate clearly impractical suggestions.

   • Identify promising concepts.

3. Refinement:

   • Develop promising ideas further.

   • Combine complementary concepts.

   • Add details and specifications.

3. Decision Matrix and Decision Tree

Decision Matrix (Pugh Matrix)

1. Structure:

   • Rows: Alternative solutions.

   • Columns: Evaluation criteria.

   • Cells: Scores/ratings for each alternative against each criterion.

2. Construction Steps: a. List Alternatives: All potential solutions. b. Identify Criteria: Relevant factors for decision-making. c. Weight Criteria: Assign importance weights (sum to 1 or 100%). d. Establish Baseline: Select reference alternative (often current state). e. Score Alternatives: Rate each alternative against criteria. f. Calculate Weighted Scores: Score × Weight for each cell. g. Sum and Compare: Total weighted scores for each alternative.

3. Scoring Methods:

   • Relative Scoring: + (better), 0 (same), - (worse) relative to baseline.

   • Numerical Scales: 1-5, 1-10, or 0-100 scales.

   • Pairwise Comparison: Comparing alternatives two at a time.

4. Example Criteria Categories:

   • Technical: Performance, reliability, safety.

   • Economic: Cost, ROI, payback period.

   • Implementation: Time, complexity, resources.

   • Risk: Probability of failure, consequences.

Decision Tree Analysis

1. Basic Structure:

   • Decision Nodes: Squares (choices under decision-maker's control).

   • Chance Nodes: Circles (probabilistic outcomes).

   • End Nodes: Triangles or endpoints (final outcomes with values).

   • Branches: Lines connecting nodes.

2. Construction Process: a. Identify Decision Points: Key choices to be made. b. Identify Uncertain Events: Probabilistic outcomes. c. Determine Probabilities: Estimate likelihoods for chance events. d. Assign Values: Numerical outcomes (costs, benefits). e. Calculate Expected Values: Working backward from endpoints.

3. Expected Monetary Value (EMV) Calculation:

   • For chance nodes: $EMV = \sum (Probability \times Value)$

   • For decision nodes: Choose branch with highest EMV.

   • Example: $EMV = (0.7 \times \$100,000) + (0.3 \times -\$20,000) = \$64,000$

4. Types of Decision Trees:

   • Single-Stage: One decision point.

   • Multi-Stage: Sequence of decisions over time.

   • With and Without Probabilities: Deterministic vs stochastic.

5. Sensitivity Analysis:

   • Testing how changes in probabilities or values affect decision.

   • Identifying critical assumptions.

   • Determining decision robustness.

Comparison and Application

1. When to Use Each Tool:

   • Decision Matrix: Best for multi-criteria decisions with multiple alternatives.

   • Decision Tree: Best for sequential decisions with uncertainty.

2. Limitations:

   • Matrix: May oversimplify complex trade-offs.

   • Tree: Becomes unwieldy with many branches.

   • Both: Depend on quality of input data and assumptions.

3. Combined Approach:

   • Use matrix to evaluate alternatives.

   • Use tree to analyze implementation paths.

   • Iterate between tools for complex decisions.