Morphological Matrix, then Scoring¶

A combinatorial design tool — instantiates Divergence-Convergence Cycle Orchestration

Decomposes a design into independent parameters, enumerates the grid of parameter-value combinations to force complete coverage, then scores the viable combinations — so generation is exhaustive before selection begins.

Morphological Matrix, then Scoring is the one mechanism here that spans both phases in a single artifact: it generates by combination and then selects by scoring. You break the design into a handful of independent parameters, list the possible values for each, and lay them out as a grid; every path that picks one value per parameter is a candidate configuration. Because the grid is exhaustive, it surfaces combinations no free brainstorm would think to try — including the deliberately odd ones — before any judgment is applied. Its defining move is systematic combinatorial coverage first, scoring second: breadth is guaranteed by construction rather than by inspiration, and only the surviving, coherent combinations are then rated against criteria.

Example¶

A team designing a small consumer drone decomposes it into four parameters: rotor count (3 / 4 / 6), power (swappable battery / fixed battery / tethered), camera mount (fixed / single-gimbal / dual-gimbal), and control link (phone / dedicated controller / autonomous waypoint). The grid holds 3×3×3×3 = 81 combinations. Most free-form ideation would have circled the familiar "quadcopter, fixed battery, gimbal, controller"; the matrix forces attention onto cells like hexrotor + swappable battery + dual-gimbal + autonomous that nobody proposed but that fits the inspection-market use case.

Then comes the pruning and scoring. A cross-consistency pass removes incoherent cells — tethered power with autonomous waypoint flight is contradictory, so those combinations are struck out, collapsing 81 to about 20 viable configurations. The survivors are scored against weighted criteria (flight time, payload, cost, regulatory ease), and the top three go forward. The tool produced both the option set and its first ranking.

How it works¶

What distinguishes it from either a pure generator or a pure scorer:

Decompose into independent parameters. The design is split into axes chosen to vary as independently as possible; this is the step that determines whether the grid is meaningful.
Enumerate the full grid. Every combination of one value per parameter is a candidate, so coverage is complete by construction — the odd cells are features, not noise.
Prune for consistency. Physically or logically incompatible combinations are struck out (cross-consistency assessment) before anything is scored, collapsing the grid to viable configurations.
Score the survivors. Only coherent combinations are rated against weighted criteria, yielding a ranking — generation and selection completed in one artifact.

Tuning parameters¶

Parameter granularity — how finely the design is decomposed and how many values per axis. Finer grids cover more but explode combinatorially; coarser grids stay tractable but may miss configurations.
Independence of axes — how orthogonal the chosen parameters are. Well-separated axes make the grid meaningful; overlapping ones generate redundant or contradictory cells and inflate the pruning burden.
Consistency-pruning strictness — how aggressively incoherent combinations are removed before scoring. Strict pruning keeps the scored set small and sane; loose pruning preserves surprising hybrids at the cost of noise.
Scoring weight scheme — the criteria and their weights applied to survivors; the same dial as any weighted scorecard, and the same false-precision risk.

When it helps, and when it misleads¶

Its strength is forced coverage: derived from Fritz Zwicky's morphological analysis, it guarantees the option space is explored systematically rather than left to whoever speaks first, and it routinely surfaces viable configurations a free-form session never reaches.^[1] Pairing enumeration with immediate consistency-pruning and scoring makes it a compact bridge from breadth to a shortlist.

Its signature failure is combinatorial explosion: add a few parameters or values and the grid grows past comprehension, so the tool becomes unusable exactly when the design is rich. It also assumes the parameters are genuinely independent — when they aren't, the grid fills with contradictory cells and the pruning step does most of the real work. And the final scoring inherits every hazard of weighted scorecards, including false precision and being run backwards to bless a favored cell. The discipline that keeps it useful is choosing few, truly independent axes; pruning hard for consistency before scoring; and treating the grid as a coverage aid, not proof that the best design must be one of its cells.

How it implements the components¶

Uniquely among the set, it fills components on both sides of the boundary within one tool:

exploration_dimension_map — the parameters and their value lists are the map of the space the tool explores; the axes define the exploration directly.
candidate_option_inventory — every coherent grid cell is an enumerated, provenance-tagged candidate configuration, giving a complete inventory rather than a sampled one.
option_comparison_frame — the scoring pass rates the surviving combinations against common criteria, producing a comparison across the viable set.
convergence_criterion_set — the weighted criteria applied to survivors are the declared basis for narrowing the grid to a shortlist.

It does not run the human process around the analysis — it neither enforces a divergence_phase_contract nor writes the durable convergence_commitment_record; those belong to Divergence-Convergence Workshop and Multi-Criteria Decision Matrix.

Instantiates: Divergence-Convergence Cycle Orchestration — it is the systematic generator-plus-scorer that spans one full diverge-then-converge beat.
Consumes: nothing upstream — it generates its own candidates from the parameter decomposition.
Sibling mechanisms: Assumption Reversal and Recombination is the intuitive counterpart generator · Multi-Criteria Decision Matrix shares its scoring half but supplies no generation · Concept Screening Funnel · Double Diamond Workflow · Divergence-Convergence Workshop · Design Studio Critique Cycle · Dot Voting with Evidence Gate

References¶

[1] Morphological analysis — the systematic charting of a problem's parameters and their possible values to enumerate all combinations — was developed by astronomer Fritz Zwicky; cross-consistency assessment, the pruning of incompatible combinations, is part of the same method and is why the scoring step here operates only on coherent cells. ↩