Skills / Innovation-creativity

TRIZ

An ideal machine is one that does not exist, yet performs its function.

TRIZ (Theory of Inventive Problem Solving) is a systematic innovation methodology developed by Soviet engineer Genrich Altshuller after analyzing patterns across hundreds of thousands of patents. Rather than relying on brainstorming or trial-and-error, TRIZ gives innovators a structured path: model any design challenge as a contradiction, then look up which of 40 Inventive Principles have historically resolved that class of conflict. It is the engineering world's most rigorous answer to the question: why guess when invention has already been catalogued?

By Genrich Altshuller  ·  Free

Use this skill See it in action

Specimen 01 · Live diagnosisTRIZ

Input

“We're designing a thermal management system for a high-power battery pack. We need thicker heat-sink fins to dissipate more heat, but thicker fins…”

Diagnosis

TRIZ starts with contradictions, not solutions.

Full transcript ↓

Calibrated referenceagent-skills.ai

The gap

Resolve design contradictions using patterns from 400,000 analyzed patents

TRIZ is built on a single insight: genuine inventions resolve contradictions rather than accept trade-offs. Altshuller identified two types — Technical Contradictions (improving parameter A degrades parameter B, e.g., making a structure stronger makes it heavier) and Physical Contradictions (one element must simultaneously have opposite properties). Problems are expressed using the 39 Engineering Parameters, a standardized vocabulary of system attributes. The 39×39 Contradiction Matrix then maps each parameter conflict to the 3–5 Inventive Principles most often used to resolve it in patent history. The 40 Inventive Principles (e.g., Segmentation, Prior Action, The Other Way Round, Phase Transitions) are concrete, domain-independent solution strategies with worked examples. The Ideal Final Result (IFR) concept anchors all exploration: the ideal system performs its function without existing — no added complexity, cost, or harm. For physical contradictions, four Separation Principles (in time, in space, in condition, between whole and parts) guide resolution. Advanced work uses ARIZ (an 85-step algorithm for intractable problems), 76 Inventive Standards via Substance-Field analysis, and Trends of Engineering System Evolution for technology forecasting.

The problem

Engineers and product designers routinely hit the same wall: improving one performance parameter degrades another, and every proposed fix is a compromise. Conventional brainstorming recycles familiar solutions; benchmarking finds no analogues; the team debates trade-offs indefinitely. TRIZ addresses this directly — Altshuller showed that 'can't have it both ways' is not a law of physics but a signal that the right inventive principle has not yet been applied.

The solution

Stop compromising on design trade-offs. Model your problem as a contradiction, look up which inventive patterns have already resolved it across thousands of analogous patents, and leave the conversation with a concrete shortlist of solution directions no amount of brainstorming would have surfaced.

You bring

A specific design problem or engineering trade-off you are stuck on
A description of what you are trying to improve and what gets worse when you do
The domain context (mechanical, software, process, business) so principles can be adapted
Willingness to describe the problem in terms of system components and interactions

You get

A formally modeled contradiction using the 39 Engineering Parameters
A ranked shortlist of 3–5 Inventive Principles from the Contradiction Matrix matched to your specific parameter conflict
Concrete adaptations of each principle to your context with analogous examples from other industries
An Ideal Final Result statement that reframes what a genuine solution must achieve
A written Contradiction Analysis document capturing the full problem model and solution directions

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.01  —  The same situation, two ways

Specimens A · B

Specimen AVague

Unaided

“Your R&D team has spent three months on a drone propulsion system. The motor needs a larger rotor for efficiency, but a larger rotor increases moment of inertia, slowing response time, which is the spec you cannot sacrifice. Every design iteration trades one requirement against the other. The project review deck has seven slide versions of the same bar chart showing the same inverse relationship. No one has a path forward.”

×Vague — nothing concrete to act on.

Specimen BCalibrated

With TRIZ

“In a single TRIZ session you model the contradiction: improving Parameter 18 (Brightness of Light / Energy Conversion Efficiency, proxied for propulsive efficiency) worsens Parameter 9 (Speed) via inertia. The Contradiction Matrix returns Principles 8 (Anti-Weight), 15 (Dynamics), 26 (Copying), and 35 (Parameter Changes). Principle 15 (Dynamics) immediately surfaces the counter-rotating dual-rotor concept — two smaller, lighter, counter-spinning rotors achieve the efficiency of a large rotor while their opposing angular momenta cancel, reducing net inertia. The IFR statement ('the rotor itself generates maximum thrust with zero rotational inertia') confirms this is a level-3 solution. You leave the session with a documented contradiction analysis, the Matrix result card, and three concrete concept directions to prototype.”

◦Specific, grounded, and ends with one clear next step.

Fig.02  —  A real session, unedited

In-app · assistant

TRIZClaude

TRIZ

TRIZ starts with contradictions, not solutions. Tell me the specific trade-off you're stuck on — the thing you want to improve, and what gets worse when you try. It doesn't need to be in engineering language yet; describe it the way you'd explain it to a colleague.

We're designing a thermal management system for a high-power battery pack. We need thicker heat-sink fins to dissipate more heat, but thicker fins mean less room for more fins, which reduces total surface area. It's a circle we can't get out of.

That's a clean Technical Contradiction. Improving Parameter 14 (Strength / thermal conductivity via fin thickness) worsens Parameter 7 (Volume of Stationary Object — the inter-fin space that accommodates cooling flow and total fin count). Before I look up the Contradiction Matrix, let me check: does the fin material also need contradictory properties at the same time — for example, must it be both rigid and flexible, or both thick and thin simultaneously? If so, we may also have a Physical Contradiction, which uses a different TRIZ tool.

There's actually a secondary issue — the fins need to be stiff enough to survive vibration but flexible enough to accommodate thermal expansion without cracking. So yes, both rigid and flexible.

Good — you have both types. For the Physical Contradiction (rigid AND flexible), we'll apply Separation in Condition: the fin behaves rigidly under vibration (high-frequency, low-amplitude load) and flexibly under thermal expansion (low-frequency, high-amplitude displacement). This points toward composite or bi-material structures, living hinges, or geometry-based compliance — all well-documented TRIZ solutions for this separation class.

For the Technical Contradiction on thickness vs. surface area, the Contradiction Matrix maps Parameter 14 (Strength) vs. Parameter 7 (Volume of Stationary Object) to Inventive Principles 1, 13, 17, and 40:

- Principle 1 (Segmentation): Divide the fin into smaller independent elements — micro-fin arrays or pin-fin geometries achieve the same or greater surface area at reduced individual-fin thickness.- Principle 13 (The Other Way Round): Instead of thickening the fin to improve conductance, invert the problem — use a phase-change or vapor-chamber structure embedded in a thin fin that moves heat laterally rather than relying on conduction depth.- Principle 17 (Another Dimension): Move fins into a 3D geometry — offset, staggered, or folded fin structures increase surface area without consuming volume in only one plane.- Principle 40 (Composite Materials): Use a graded-conductivity material — high conductivity at the base, lower-density structure at the tip — so heat capacity per unit volume increases without requiring uniform thickness.

Principle 13 is interesting — we hadn't considered vapor chambers at all. Can you help frame what the Ideal Final Result would look like here so we can use it as a filter?

Here is your IFR statement in TRIZ format:

'The thermal management structure itself dissipates the full heat load to the ambient with zero increase in volume, zero added weight, and zero mechanical stress, without requiring any active components or external intervention.'

This tells you exactly what to filter for: any solution that adds volume, weight, or a pump/fan moves you away from IFR. Vapor chambers score well — they move heat with minimal volume by exploiting phase change rather than conduction geometry. The IFR also suggests a direction Principle 13 didn't make explicit: look for heat paths that exploit an already-present resource in the system — the battery electrolyte, the structural casing, the existing air gap — rather than adding new heat-sink material at all.

That last point about using the structural casing as a heat path — that hadn't occurred to anyone on the team. This is already more structured than six months of internal review.

Reply to TRIZ…

What the skill is doing

01Contradiction Classification

02Parameter Mapping

03Contradiction Matrix Lookup

Fig.03  —  The same skill, in the channel you already use

WhatsApp · live

9:41

TTRIZtyping…

Messages are end-to-end encrypted. No one outside of this chat can read them.

Today

We're designing a thermal management system for a high-power battery pack. We need thicker heat-sink fins to dissipate more heat, but thicker fins mea…9:14✓✓

TRIZ starts with contradictions, not solutions.9:14

Tell me the specific trade-off you're stuck on — the thing you want to improve, and what gets worse when you try.9:15

It doesn't need to be in engineering language yet; describe it the way you'd explain it to a colleague.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

Contradiction Classification

Determines whether the user's problem is a Technical Contradiction (improving A worsens B), a Physical Contradiction (same element needs opposite properties), or both. This classification decides which TRIZ tools apply and prevents wasted effort on wrong-level solutions.

◦ MethodBased on Altshuller's foundational distinction between the two contradiction types, which require different resolution strategies — the Contradiction Matrix for technical contradictions and the four Separation Principles for physical contradictions.

CapabilityC-02

Parameter Mapping

Translates the user's real-world problem language into TRIZ's standardized vocabulary of 39 Engineering Parameters (e.g., 'too heavy' → Parameter 1: Weight of Moving Object; 'too slow' → Parameter 9: Speed). This translation is the hardest step for new TRIZ users and the one that most determines result quality.

◦ MethodThe 39 Engineering Parameters are Altshuller's distilled vocabulary from patent analysis — a domain-neutral language that lets any technical conflict be expressed in terms comparable to historical patent solutions.

CapabilityC-03

Contradiction Matrix Lookup

Once the improving and worsening parameters are identified, consults the 39×39 Contradiction Matrix to retrieve the specific Inventive Principles most frequently used to resolve that exact parameter conflict across thousands of patents. Presents the top 3–5 principles with their frequency ranking.

◦ MethodThe Altshuller Matrix (updated as Matrix 2003 by Darrell Mann with 48 parameters and 4,000+ data points) is a direct lookup table derived from patent analysis — each cell represents empirical frequency, not opinion.

CapabilityC-04

Inventive Principle Coaching

Explains each suggested Inventive Principle concretely (e.g., Principle 1 Segmentation, Principle 13 The Other Way Round, Principle 35 Parameter Changes) with worked examples from industries outside the user's domain, then co-develops how that principle could apply to the user's specific system.

◦ MethodAltshuller's 40 Inventive Principles are domain-independent solution patterns; the skill's role is to bridge the gap between the generic principle description and specific application — the most common failure point when TRIZ is used without a coach.

CapabilityC-05

Ideal Final Result Formulation

Helps the user write a precise IFR statement — what the system would do if it solved itself, with no added components, cost, or side effects. The IFR is then used as a filter to evaluate candidate solutions and prevent over-engineering.

◦ MethodThe Ideal Final Result is a core TRIZ concept: 'An ideal machine is one that does not exist, yet performs its function' (Altshuller). IFR shifts thinking from 'how can I fix this?' to 'what does a complete solution look like?' — often generating level-3 and level-4 innovations.

CapabilityC-06

Physical Contradiction Separation

When the same system element must simultaneously satisfy contradictory requirements, applies the four Separation Principles: Separation in Time (property A at time T1, property B at time T2), Separation in Space (A in location 1, B in location 2), Separation on Condition (A when condition X, B when condition Y), and Separation Between Whole and Parts.

◦ MethodPhysical contradictions cannot be resolved by the Contradiction Matrix; Altshuller developed the four Separation Principles specifically for cases where compromise would destroy the required function.

CapabilityC-07

Levels of Innovation Assessment

Helps the user understand which of Altshuller's five innovation levels their problem requires — from Level 1 (routine design within current knowledge) to Level 5 (discovery requiring new scientific knowledge) — calibrating how far outside current domain knowledge the search for a solution must go.

◦ MethodAltshuller's 5 Levels of Invention, derived from patent analysis, show that 77% of patents are Levels 1–2 (solvable with known-domain knowledge) while Levels 3–5 require cross-domain analogy — exactly what the Contradiction Matrix and Inventive Principles enable.

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

Contradiction Analysis Document

A structured one-page writeup that formalizes the problem: the undesired effect, the identified technical contradiction expressed as improving/worsening parameter pairs, and the physical contradiction if present. Serves as the sharable foundation for team solution sessions.

OutputD-02

Contradiction Matrix Result Card

A formatted output showing the selected parameter pair, the matching cell in the Altshuller Matrix, and the 3–5 suggested Inventive Principles with their numbers, names, and definitions. Functions as a reference card throughout the solution development process.

OutputD-03

Ideal Final Result Statement

A precise, structured IFR statement written in TRIZ format: '[The system element] itself [performs the desired function] [without any negative effects] [without any added complexity or cost].' Used to evaluate and filter candidate solutions.

OutputD-04

Principle Adaptation Worksheet

For each of the top 3 suggested Inventive Principles: the principle name and number, one or two analogue examples from other industries, and a set of brainstorming prompts specifically tailored to applying that principle to the user's system.

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

Genrich Altshuller

Genrich Altshuller (1926–1998) was a Soviet engineer, inventor, and science-fiction author who founded the field of systematic innovation. Beginning in 1946, he spent decades analyzing patent literature — eventually more than 400,000 patents across every engineering domain — to identify the repeating patterns underlying genuine invention. He developed TRIZ over 50 years of research, established MATRIZ (the International TRIZ Association), and authored foundational texts including 'The Innovation Algorithm,' 'Creativity as an Exact Science,' and 'The Art of Inventing.' Today TRIZ is institutionalized at Samsung, Boeing, Intel, P&G, Volkswagen, and hundreds of universities worldwide.

Status · Inspired by Genrich Altshuller’s work — not yet claimed. Are you Genrich Altshuller?

Primary sourceS-01

The Innovation Algorithm (Algoritm Izobreteniya)

by Genrich Altshuller

Soviet engineer and patent examiner; founder of TRIZ; author of 8 foundational TRIZ texts; established MATRIZ certification body; analyzed 400,000+ patents over 50 years.

Read the original ↗

Citationaltshuller.ru

In the build queue

Be first to run it.

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

Notify meEmail

At launchI have a design trade-off where improving [X] always makes [Y] worse and we've been stuck going in circles for weeks. Can we model it as a TRIZ contradiction and look up what the patent literature suggests?