Sign inGet started
Skills / Problem-solving

Morphological Analysis

The morphological method is concerned with the inventing, ordering, and understanding of all possible solutions to multi-dimensional problems.

General Morphological Analysis (GMA) is Fritz Zwicky's systematic method for exhaustively mapping the complete solution space of any complex, multi-dimensional problem — then filtering it to what is actually feasible. By constructing a Zwicky Box matrix (parameters × variants) and applying Cross-Consistency Assessment, practitioners surface solution combinations that would never emerge from brainstorming alone. Designed for systems engineers, strategic planners, and innovation teams wrestling with problems that have too many interdependent variables to explore intuitively.

By Fritz Zwicky · Free
Specimen 01 · Live diagnosisMorphological Analysis
Input

“We're designing a remote monitoring system for industrial equipment. We've been debating the same three architectures for six weeks and can't decide.”

Diagnosis
I use General Morphological Analysis — Fritz Zwicky's method for mapping the complete solution space of a complex problem, then filtering it to what is actually feasible.
Full transcript ↓
Calibrated referenceagent-skills.ai
The gap

Map every possible solution, then eliminate what can't work

General Morphological Analysis follows a 7-step sequence: (1) define and delimit the problem precisely; (2) identify all key parameters — the independent dimensions that together characterize the problem space; (3) enumerate all conceivable variants for each parameter, without judgment; (4) construct the Zwicky Box — a matrix where rows are parameters and columns are their variants; (5) survey the resulting combination space; (6) apply Cross-Consistency Assessment (CCA), checking every pair of parameter variants for logical compatibility and eliminating contradictory combinations; (7) evaluate the surviving solution field and elaborate the most promising configurations. A 5-parameter problem with 4 variants each yields 1,024 theoretical combinations; CCA typically reduces this to a manageable solution field of dozens. The methodology is domain-agnostic: Zwicky applied it to galaxy cataloguing, jet engine design, and policy forecasting within the same decade.

The problem

Complex design problems have an enormous but finite solution space — yet most problem-solving approaches explore only a narrow slice of it, anchored around familiar options. Teams default to debating the same two or three architectures, strategies, or configurations they already know, never checking whether there are feasible combinations they have never considered. This availability bias masquerades as exhaustive analysis.

The solution

Walk away with a complete Zwicky Box for your problem — a visual matrix that makes the entire solution space visible — and a filtered set of internally consistent combinations identified through Cross-Consistency Assessment, so your shortlist is bounded by logic rather than habit.

You bring
  • A complex problem with multiple interacting design or strategy dimensions
  • Enough domain knowledge to validate parameter choices — or willingness to build this collaboratively
  • A specific decision context: product design, policy, technology architecture, or scenario planning
  • Openness to discovering solution combinations that were not on your original shortlist
You get
  • A completed Zwicky Box matrix mapping all parameters to their full set of conceivable variants
  • A Cross-Consistency Assessment showing which variant combinations are logically compatible
  • A filtered solution field of internally consistent candidate configurations
  • Narrative elaborations of the most promising combinations, ready for evaluation or prototyping
In action

Watch the methodology work.

Three specimens from a single real session: the same situation, unaided and calibrated, the full transcript, and the skill answering live in the channel where the work happens.

Fig.01The same situation, two ways
Specimens A · B
Specimen AVague
Unaided

“Your team has been debating three monitoring architectures for six weeks — cloud-based, edge gateway, and hybrid. The conversation keeps looping. No one has verified whether these three are internally consistent across all design dimensions, or whether there are architectures outside your current frame that deserve consideration.”

×Vague — nothing concrete to act on.
Specimen BCalibrated
With Morphological Analysis

“You have a Zwicky Box with 6 parameters and 24 variants, representing 4,096 theoretical configurations. After Cross-Consistency Assessment, 11 feasible solution clusters survive. Three are genuinely novel — including an energy-harvesting combined with on-sensor processing architecture your team never surfaced. Two of your original three architectures passed the CCA filter. One contained a hidden internal contradiction.”

Specific, grounded, and ends with one clear next step.
Fig.02A real session, unedited
In-app · assistant
Morphological AnalysisClaude
Morphological Analysis
I use General Morphological Analysis — Fritz Zwicky's method for mapping the complete solution space of a complex problem, then filtering it to what is actually feasible. Before we build the Zwicky Box, I need to delimit the problem precisely. What are you designing, deciding, or solving?
We're designing a remote monitoring system for industrial equipment. We've been debating the same three architectures for six weeks and can't decide.
That's a classic sign you've anchored on familiar options instead of mapping the full solution space. Let's fix that. First, delimitation: is this a retrofit to existing machines, or can you modify the equipment itself? And what's the deployment scale — dozens of sensors or thousands?
Retrofit only — no changes to existing machines. Small-to-mid plants, so roughly 50 to 500 sensors per deployment.
Good. Now parameter elicitation — the independent dimensions that together define any possible design. Here is a starting set:
1.
Data transmission protocol — cellular LTE, industrial WiFi, wired Ethernet, LPWAN (LoRa)2. Processing architecture — cloud-only, edge gateway, hybrid edge-cloud, on-sensor3. Data model — time-series database, relational, graph, event-stream4. Alert mechanism — threshold rules, ML anomaly detection, physics-based modeling, operator-defined5. Integration layer — REST API, MQTT broker, OPC-UA, proprietary
Each becomes a row in the Zwicky Box. Are any critical dimensions missing, or should we replace one?
I'd add 'power source' as a dimension — some sensors in our environments can't be hard-wired to mains power.
Essential addition. Power Source variants: wired mains, battery (replaceable), energy harvesting (vibration/thermal), Power over Ethernet (PoE). That gives us 6 parameters. With 4 variants each, the theoretical box contains 4⁶ = 4,096 combinations. This is exactly where Cross-Consistency Assessment earns its value — we systematically identify which variant pairings are logically contradictory. For example: 'edge-only processing' paired with a 'cloud-only data model' is internally inconsistent. Eliminating that pairing removes an entire slice of configurations simultaneously. Want to start the CCA pass?
Yes — and I'm already noticing we never considered energy harvesting combined with on-sensor processing. That might actually work for our outdoor installations.
That is exactly what the Zwicky Box surfaces: combinations that do not appear on anyone's original shortlist. Let's run the CCA and see which configurations survive — then I'll elaborate the most promising clusters into full solution descriptions you can take to your architecture review.
Reply to Morphological Analysis
What the skill is doing
01Problem Delimitation
02Parameter Elicitation
03Zwicky Box Construction
Fig.03The same skill, in the channel you already use
WhatsApp · live
9:41
MAMorphological Analysistyping…
Messages are end-to-end encrypted. No one outside of this chat can read them.
Today
We're designing a remote monitoring system for industrial equipment. We've been debating the same three architectures for six weeks and can't decide.9:14✓✓
I use General Morphological Analysis — Fritz Zwicky's method for mapping the complete solution space of a complex problem, then fi…9:14
Before we build the Zwicky Box, I need to delimit the problem precisely.9:15
What are you designing, deciding, or solving?9:15
Message

The same skill, where the work happens.

No new app to learn. The methodology runs over the WhatsApp Business API, so the answer lands as a reply in the thread you’re already in — same rigour, zero context-switch.

Reads the situation, names the pattern, returns one concrete next move.
Delivered in seconds, inside a conversation that already exists.
Specimen · WhatsApp Business API · live
Capabilities

What it does, specifically.

Each capability is a distinct move drawn straight from the source methodology — not a generic assistant guessing.

CapabilityC-01

Problem Delimitation

Guides you to define and bound the problem precisely before constructing the box. Clear delimitation prevents parameter drift and keeps the Zwicky Box from expanding to an unmanageable size. This step distinguishes GMA from open-ended brainstorming: the boundaries you set here define the scope of the entire analysis.

Based on Zwicky's first step: precise problem definition is a prerequisite to parameter selection — without it, the resulting box is undefined and the CCA filter cannot be applied consistently.
CapabilityC-02

Parameter Elicitation

Surfaces the essential dimensions of your problem through structured questioning. Parameters must be independent of each other, collectively sufficient to characterize any possible solution, and genuinely relevant to the problem space — not just the attributes that first come to mind.

Mirrors Zwicky's second step — identifying parameters (dimensions/attributes) that together fully characterize the problem space; Zwicky identified parameter selection as the hardest and most expert-dependent step in GMA.
CapabilityC-03

Zwicky Box Construction

Builds the signature matrix artifact: rows are parameters, columns are all conceivable variants for each parameter. The resulting box makes visible the complete solution space, including combinations that would never surface through normal ideation — many of them novel and some of them highly promising.

The Zwicky Box (morphological box) is the central artifact of GMA; for a 5-parameter × 4-variant problem it contains 1,024 theoretical configurations, each representing a distinct and complete solution candidate.
CapabilityC-04

Cross-Consistency Assessment

Walks you through pairwise compatibility checks across all combinations of parameter variants. Internally contradictory pairings are identified and removed, collapsing a potentially enormous theoretical space into a tractable solution field of logically coherent candidates.

CCA is the analytical engine that distinguishes GMA from simpler morphological charts; it applies pairwise compatibility logic — if variant A1 is incompatible with variant B3, all configurations containing both are eliminated simultaneously.
CapabilityC-05

Solution Field Analysis

Identifies the most coherent and promising clusters within the filtered solution field. Each surviving configuration is elaborated into a narrative solution description — all parameters specified, internal logic explained — suitable for comparative evaluation or immediate development.

Corresponds to Zwicky's evaluation step and Ritchey's 'solution cluster' methodology developed at the Swedish Defence Research Agency (FOI) for applying GMA to wicked, non-quantified societal problems.
Tested

Graded before it shipped.

Every skill is scored against independent scenarios for methodology fidelity before it goes live — not vibes, a rubric.

What it produces
OutputD-01

Zwicky Box Matrix

The master parameter × variant matrix showing the complete solution space for your problem. Each row is a problem parameter; each column is a conceivable variant. Reading across the columns, one choice per row, traces any single complete solution configuration.

OutputD-02

Cross-Consistency Compatibility Map

A pairwise comparison grid showing which combinations of parameter variants are logically compatible and which are contradictory. Each marked contradiction eliminates an entire slice of configurations from the solution space simultaneously.

OutputD-03

Solution Field Report

The filtered set of internally consistent configurations surviving CCA, with narrative elaborations of the most promising combination clusters. Each entry specifies all parameters and explains the internal coherence of the configuration.

The source

Grounded in the original work.

Every answer traces back to a real source and the practitioner who wrote it — not a secondhand summary. Here is the source of record.

Source authorA-01

Fritz Zwicky

Fritz Zwicky (1898–1974) was a Swiss-American astrophysicist at the California Institute of Technology (Caltech) who developed General Morphological Analysis in the 1940s and formalized it in his 1969 book. He predicted the existence of neutron stars, coined the term 'dark matter,' and catalogued over 10,000 galaxies — applying the morphological method across astrophysics, WWII-era jet propulsion engineering at Aerojet, and scientific theory. His method has since been extended computationally by Tom Ritchey at Sweden's Defence Research Agency (FOI) for application to complex policy and societal challenges.

Status · Inspired by Fritz Zwicky’s work — not yet claimed. Are you Fritz Zwicky?
Primary sourceS-01

Discovery, Invention, Research Through the Morphological Approach

by Fritz Zwicky

Swiss-American astrophysicist, Caltech professor (1942–1968), predictor of neutron stars and dark matter, Aerojet co-founder, developer of GMA applied to engineering design and galaxy cataloguing.

Read the original ↗
Citationen.wikipedia.org
In the build queue

Be first to run it.

Morphological Analysis is being built right now. Leave your email and we’ll tell you the moment it goes live.

Notify meEmail
At launchI'm designing [product/system/policy] and I suspect I'm anchored on too few options. Can we build a Zwicky Box to map the full solution space?