Morphological Analysis (Morphological Box)

Overview

Invented by Fritz Zwicky. Break a problem into independent dimensions, list possible values for each dimension, then systematically generate all combinations. The structure guarantees exhaustive coverage.

Goal

Generate all possible solutions by identifying independent dimensions and systematically combining values across dimensions. Combination generation is purely mechanical.

Steps

Step 1: Define the Problem

Clearly state what you’re trying to create or solve. This frames what dimensions are relevant.

Output: Problem statement

Step 2: Identify Independent Dimensions

List the independent aspects/parameters of the solution.

Rules for good dimensions:

• Independent (changing one doesn’t force change in another)
• Relevant (affects the solution quality)
• Variable (has multiple possible values)

Common dimension types:

• Physical: size, material, shape, color
• Functional: method, mechanism, process
• Contextual: user, location, time, frequency
• Economic: price point, cost structure

Output: List of 4-8 dimensions

Step 3: List Values for Each Dimension

For each dimension, list all possible values. Be exhaustive within reason (3-7 values per dimension typical).

Include:

• Obvious values
• Extreme values
• Novel values
• “None” or “opposite” if applicable

Output: Values for each dimension

Step 4: Construct Morphological Box

Create matrix with dimensions as rows and values as columns. This visualizes the solution space.

Output: Morphological box matrix

Step 5: Calculate Combination Count

Total combinations = V1 × V2 × V3 × … × Vn where Vi is the number of values for dimension i.

If too large (>1000), either:

• Reduce values per dimension
• Use sampling instead of exhaustive
• Add constraints to eliminate combinations

Output: Combination count

Step 6: Generate Combinations

Systematically generate all combinations. Each combination picks one value from each dimension.

For small spaces: List all For large spaces: Sample systematically or use constraints

Output: List of combinations

Step 7: Apply Constraint Filter

Remove combinations that are:

• Physically impossible
• Logically contradictory
• Clearly inferior (dominated by another)
• Outside scope/budget

Output: Reduced combination list

Step 8: Evaluate Viable Combinations

Score remaining combinations on:

• Feasibility (can we build it?)
• Value (does it solve the problem well?)
• Novelty (is it differentiated?)
• Cost (can we afford it?)

Output: Ranked combinations

Step 9: Select Top Combinations

Select top 3-5 combinations for further development. Consider diversity (don’t pick all similar).

Output: Shortlist

When to Use

• Designing new products/solutions
• Exploring solution space exhaustively
• Ensuring no combinations are missed
• Breaking creative blocks
• Systematic innovation

Verification

• Dimensions are truly independent
• Values are mutually exclusive within each dimension
• Values are exhaustive (cover the space)
• Combination count is correct
• Constraints are justified
• Evaluation criteria are clear

Input: $ARGUMENTS

Apply this procedure to the input provided.