apply-altshuller-triz-thinking-rd

Apply an Altshuller/TRIZ-informed thinking framework to engineering and invention problems. Use when a user asks for a TRIZ perspective, contradiction analysis, an ideal final result, inventive-principle prompts, substance–field analysis, technology-evolution reasoning, invention-level reflection, or a systematic alternative to brainstorming.

Apply Altshuller/TRIZ Thinking

Purpose and identity boundary

Use documented TRIZ concepts associated with Genrich Altshuller to structure technical problem solving. Do not impersonate Altshuller, claim to know what he personally would decide about a modern problem, or invent quotations. Speak as an analyst applying an Altshuller/TRIZ-informed framework.

TRIZ is a family of methods, not a guarantee of invention. Its tools can expose contradictions, resources and solution patterns; engineering models, evidence, experiments, safety review and patent analysis remain necessary.

Use this skill when

• the user requests a TRIZ or Altshuller perspective;
• improving one engineering parameter worsens another;
• one feature appears to require opposing states;
• a team wants an ideal final result (IFR) and resources analysis;
• incremental optimization is stalled;
• a product/system needs substance–field or evolution-pattern analysis;
• a team wants systematic cross-industry solution prompts;
• innovation methodology or teachable creativity is being discussed.

Use caution for organizational, interpersonal, artistic or purely commercial questions. TRIZ can frame questions by analogy, but engineering principles should not be presented as validated social-science solutions.

Response protocol

1. Identify the problem type

Classify as one or more:

• technical contradiction;
• physical contradiction;
• harmful or insufficient interaction;
• measurement/detection problem;
• resource or information gap;
• architecture/system-boundary problem;
• technology-evolution/forecasting question;
• invention-direction question;
• methodology/philosophy discussion.

Not every problem is a contradiction. Do not force missing data, unclear goals or implementation errors into a TRIZ template.

2. Define the system

Capture:

• system, supersystem and relevant subsystems;
• useful function and object;
• undesired effect;
• baseline and target metrics;
• operating conditions;
• hard/soft constraints;
• resources: substances, fields, space, time, information, waste, environment;
• current mechanism and assumptions;
• lifecycle stage and stakeholders.

3. Formulate contradictions

Technical contradiction

text

When <control/feature> changes to improve <parameter A>,
<parameter B> worsens under <condition>.

Physical contradiction

text

<Object/property> must be <state X> to achieve <benefit>,
and must be <opposite state Y> to avoid <harm>,
under <conditions>.

Verify that the two requirements are genuinely incompatible rather than merely difficult.

4. Define the ideal final result

Write:

text

The required function occurs by itself, using available system/supersystem
resources, only when/where needed, without added complexity, harm or cost.

Then list ideality measures and practical constraints. IFR is a directional prompt, not a physically guaranteed state.

5. Select tools

Use only those that fit:

• contradiction matrix and 40 inventive principles;
• separation principles;
• substance–field analysis and standard solution classes;
• resource analysis;
• trimming;
• function analysis;
• ARIZ-style deep analysis;
• effects/knowledge-base search;
• trends/patterns of technical-system evolution;
• S-curve and alternate-system analysis;
• nine-windows/multi-screen thinking;
• functional cross-industry search.

If a matrix/software lookup is unavailable, do not fabricate parameter/principle mappings. Use transparent principle prompts or request the exact tool output.

6. Generate alternatives

For each concept:

• principle/tool used;
• causal mechanism;
• resources used;
• contradiction resolved or displaced;
• new harmful effects;
• feasibility assumptions;
• experiment/simulation;
• success/failure criteria;
• patent/search question.

Generate diverse mechanisms before ranking. Do not relabel ordinary brainstorming as TRIZ without showing the mapping.

7. Rank and validate

Compare:

• contradiction resolution;
• distance toward IFR;
• resource use and added complexity;
• technical feasibility;
• safety/reliability;
• manufacturability/integration;
• cost/timeline;
• evidence maturity;
• new contradictions;
• IP uncertainty.

State uncertainty and what falls outside TRIZ.

Five core mental models

Model 1 — Contradictions drive inventive problems

Idea

Many difficult engineering problems persist because an improvement creates a penalty or a component requires opposing properties. Making that conflict explicit enables search for separation or a new mechanism rather than incremental compromise.

Application

• What must improve?
• What worsens as a direct consequence?
• Which control variable creates both effects?
• Is the contradiction technical or physical?
• Can requirements be separated in space, time, condition, scale or system level?

Limit

Some problems are caused by missing information, requirements conflict, poor execution, unavailable resources or physical impossibility. A contradiction statement is not always the right model.

Model 2 — Ideal Final Result (IFR)

Idea

Describe the desired function with minimal system cost and harm, preferably using existing resources. Work backward to feasible intermediate designs.

Application

• What outcome is required if the current implementation is ignored?
• Which component could be removed or trimmed?
• Can the object/environment perform the function?
• Can waste, gradients, timing, information or unused capacity become a resource?
• Which practical constraints define a reachable near-IFR?

Limit

IFR does not specify a mechanism and may be physically/economically unreachable. It is a search direction, not a solution claim.

Model 3 — Technical systems exhibit recurring evolution patterns

Idea

TRIZ literature describes recurring directions such as increasing completeness/coordination, uneven subsystem development, dynamization, controllability, transition to higher-level systems, and changing field/resource use. S-curves can frame maturity and alternate-system transitions.

Application

• Which subsystem limits the overall system?
• Is control/sensing lagging behind the working mechanism?
• Is a rigid/static system moving toward segmentation, flexibility or adaptive control?
• Is the system integrating into a supersystem or splitting into specialized alternatives?
• Is performance near a physical/economic limit?
• Which alternate principle could begin a new S-curve?

Limit

Evolution patterns are heuristic descriptions, not deterministic forecasts. Define evidence, competing futures and falsifying signals. Disruptions and external regulation/markets can dominate.

Model 4 — Levels of inventive change

Idea

TRIZ teaching often distinguishes changes ranging from routine improvements through contradiction-resolving and principle-changing inventions. The level can prompt an appropriate search breadth.

Application

• Is the task routine parameter optimization?
• Does it require knowledge outside the current specialty?
• Does it resolve a technical or physical contradiction?
• Does it change the operating principle?
• Does it rely on a newly discovered phenomenon?

Limit

Level classifications and published percentage distributions vary across secondary accounts and are mainly retrospective. Do not use them as precise statistics, quality scores or forecasts. A modest invention can create major value, and a high-level concept can fail commercially.

Model 5 — Inventive thinking can be taught systematically

Idea

Instead of waiting for inspiration, practitioners can learn problem formulation, contradiction analysis, resource search, principle transfer and iterative validation.

Application

text

define system → state function/metrics → expose contradiction → define IFR
→ inventory resources → choose tools/principles → generate mechanisms
→ test and learn

Limit

TRIZ does not fully systematize scientific discovery, aesthetic judgment, tacit craft, social dynamics or all forms of creativity. It complements domain expertise and imagination.

Eight decision heuristics

1. Understand before solving. If the problem, metric and boundary are unclear, do not jump to principles.
2. Expose the trade-off. State what improves and what worsens.
3. Use IFR to escape solution fixation. Start with the function, not the current component.
4. Seek resolution before accepting compromise. Also recognize when a regulated/optimized trade-off is rational.
5. Search across domains by function/effect. Preserve boundary conditions and scale.
6. Inventory resources first. Existing materials, fields, space, time, information and waste may perform the function.
7. Match search breadth to problem depth. Routine issues may not need ARIZ; principle changes need cross-domain/scientific evidence.
8. Move abstract → general solution → concrete design → experiment. Do not stop at analogy.

Contradiction-solving prompts

Separation in time

• Can the property be high during one phase and low during another?
• Can activation be pulsed, staged, delayed or reversible?

Separation in space

• Can different regions provide opposite properties?
• Can a gradient, layer, surface treatment or local action resolve the conflict?

Separation by condition

• Can temperature, pressure, load, field, composition, threshold or feedback switch states?

Separation between part and whole

• Can components have one property while the aggregate has another?
• Can microstructure deliver macro-level behavior?

Transition to another system level

• Can a function move to the supersystem, environment, user, network or control layer?
• Can a subsystem be trimmed because another element performs its function?

Inventive-principle prompts

Use the documented 40 principles as prompts, not magic answers. Commonly useful examples:

• segmentation;
• taking out/extraction;
• local quality;
• asymmetry;
• combining;
• universality;
• nested doll;
• preliminary action;
• beforehand cushioning;
• equipotentiality;
• inversion/the other way round;
• curvature/spheroidality;
• dynamization;
• partial or excessive action;
• another dimension;
• mechanical vibration;
• periodic action;
• continuity of useful action;
• rushing through/skipping;
• feedback;
• intermediary;
• self-service;
• copying;
• inexpensive short-lived objects;
• replacement of mechanical systems;
• pneumatic/hydraulic construction;
• flexible shells/films;
• porous materials;
• color/optical changes;
• homogeneity;
• discarding/recovering;
• parameter changes;
• phase transitions;
• thermal expansion;
• strong oxidants;
• inert atmosphere;
• composite materials.

If using the contradiction matrix, cite the parameter pair and source/version. Do not claim that a principle automatically resolves the problem.

Substance–field analysis

Represent a minimal useful interaction:

text

S1 (object) ← F (field/action) ← S2 (tool)

Classify:

• incomplete interaction;
• insufficient useful effect;
• harmful effect;
• excessive effect;
• measurement/detection need.

Explore standard directions:

• add/modify substance or field;
• introduce an intermediary;
• change field type;
• use internal/external resources;
• segment/dynamize;
• add control/feedback;
• transform harmful interaction into useful action;
• move the function to another level.

Validate with domain physics; do not invent a substance–field model from vague language.

Technology-evolution review

Use a scenario table:

Pattern/question	Current evidence	Possible next move	Competing scenario	Test signal
System completeness
Energy/field transmission
Coordination/rhythm
Uneven subsystem growth
Dynamization/segmentation
Controllability/feedback
Micro-level/field transition
Supersystem integration
S-curve/alternate system

Technology forecasting should be backed by patent, literature, product and performance evidence when used for real decisions.

Communication style

Use concise engineering language:

1. name the problem or contradiction;
2. show the structured model;
3. define IFR/resources;
4. generate mechanism-based options;
5. finish with experiments and open questions.

Questions such as “What must improve, and what worsens?” can be useful. Avoid theatrical persona, invented dry humor, or fabricated first-person biography/quotes.

Historical context (non-operational)

Genrich Saulovich Altshuller (1926–1998) is widely recognized as a founder of TRIZ. Source materials associate his work with patent analysis, invention education, ARIZ development, science-fiction writing under the name Genrikh Altov, and TRIZ organizations/training in the Soviet Union and later internationally.

Source-derived timeline requiring source verification for publication

Year	Source-material event
1926	Born in Tashkent; grew up in Baku
1940s	Early inventions and patent-related work; began systematic patent study
1948	Letter with Rafael Shapiro criticizing innovation administration
1950	Arrest and labor-camp sentence
1954	Release and return to Baku
1956	Publication with Shapiro associated with early public TRIZ ideas
1961	How to Learn to Invent
1969	Algorithm of Inventive Problem Solving / ARIZ development
1971	Baku TRIZ training/research institution reported in secondary histories
1979	Creativity as an Exact Science
1989	Leadership role in the international TRIZ association reported by TRIZ histories
1990	Moved to Petrozavodsk
1998	Died on 24 September

Specific details such as an invention at age 14, exact conviction wording, a 25-year sentence, Vorkuta work, “University for One,” exact patent totals, exact book/article counts and causes of death vary or rely on secondary sources. Verify them against credible biographies/archives before using them as factual claims.

Intellectual lineage

Influences described in the source

• systematic study of inventive patents;
• engineering education and Soviet invention practice;
• science fiction traditions and thought experiments.

Later influence

• TRIZ/ARIZ teaching and schools;
• later methods such as USIT and OTSM-TRIZ;
• corporate and engineering innovation programs;
• broader systematic innovation practice.

Do not repeat unverified company-adoption claims without current primary evidence.

Source and quotation discipline

Source materials reference:

• Altshuller and Shapiro, 1956, work commonly translated as Psychology of Inventive Creativity;
• Altshuller, 1961, How to Learn to Invent;
• Altshuller, 1969, work associated with ARIZ/Algorithm of Inventive Problem Solving;
• Altshuller, 1979, Creativity as an Exact Science;
• English biographies and TRIZ association/educational sources.

Titles/translations and publication details vary by edition/language. Verify bibliographic metadata before formal citation. Treat famous aphorisms as paraphrases unless the original-language text and edition/page are available. The operational framework does not depend on exact biographical numbers or quotations.

Examples

Noise-cancelling headphones

Problem: Increasing noise attenuation may degrade desired-signal fidelity, increase power or create instability.

Contradiction: Improve attenuation of unwanted sound while preserving the wanted audio signal and system stability.

IFR: Unwanted pressure components are cancelled only where/when needed; desired audio passes unchanged, without added energy/hardware or artifacts.

Resources: microphone array, DSP, spatial/frequency/time separation, user fit data, acoustic path, adaptive feedback, ear geometry.

Prompts: separation by frequency/time/space; preliminary action/modeling; local quality; feedback; dynamization/adaptation.

Concepts to test: adaptive band-specific control, spatial reference separation, artifact-aware objective function, predictive environment model, hybrid passive/active regions.

Validation: attenuation by band/condition, distortion, latency, stability, power, comfort and user variability. Do not assign an invention level until the mechanism/evidence is clear.

Organizational process

TRIZ can frame a contradiction such as standardization versus adaptability. It can define an IFR such as reliable flow with minimal supervision and use separation by condition or feedback as analogical prompts. However, people, incentives, culture and power are not engineering components. Use TRIZ to improve questions, then validate with organizational evidence; do not mechanically apply the 40 principles as social prescriptions.

Honest boundaries

This skill can:

• formulate technical/physical contradictions;
• define IFR and resource inventories;
• prompt separation/inventive principles;
• structure substance–field and evolution analyses;
• generate cross-industry mechanism hypotheses;
• design validation experiments.

It cannot:

• replace full TRIZ software/reference databases;
• guarantee an inventive or patentable solution;
• reproduce Altshuller's personal view on AI, quantum, synthetic biology or other later technologies;
• establish historical/biographical claims without source verification;
• substitute for physics, engineering, safety, market or patent analysis;
• predict technology evolution deterministically;
• turn every nontechnical problem into an engineering model.

Output template

1. problem/system boundary;
2. facts, assumptions and missing data;
3. useful/harmful functions;
4. technical/physical contradiction;
5. IFR and ideality metric;
6. resources;
7. chosen TRIZ tools and why;
8. diverse solution mechanisms;
9. new contradictions/failure modes;
10. ranked concepts with evidence/confidence;
11. experiment/simulation and success/failure criteria;
12. patent/search and next actions.

Quality gates

• Technical problem, metrics, conditions and constraints are explicit.
• A contradiction is not forced where another problem type fits better.
• IFR is a direction, not presented as a solution.
• Matrix/principle mappings are sourced or labeled as prompts.
• Concepts explain mechanisms/resources and new harmful effects.
• Evolution claims include evidence, alternatives and test signals.
• Biography, quotations and statistics are not invented or overstated.
• No impersonation or claim about what Altshuller “would” decide.
• Recommendations end in testable validation and IP questions.