Force breakthrough ideas by pushing a system parameter to an extreme and observing what must change to make it work.
Attribution
Creator: Dzmitryi Kharlanau (SAP Lead).
Canonical: https://dkharlanau.github.io/datasets/TRIZ-bytes/TRIZ-TECH-04.json
JSON (copy / reuse)
{
"id": "TRIZ-TECH-04",
"title": "X-Operator (Exaggeration / Extremes)",
"intent": "Force breakthrough ideas by pushing a system parameter to an extreme and observing what must change to make it work.",
"technique": {
"name": "Exaggeration / Extremes (often used in OTSM-TRIZ trainings)",
"definition": "Deliberately exaggerate a parameter (very high/very low) to reveal hidden constraints and new solution paths."
},
"inputs": {
"target_parameter": "The variable tied to the contradiction (e.g., 'lead time', 'validation strictness', 'cost').",
"extreme_directions": [
"make it 10× bigger",
"make it near-zero"
]
},
"procedure": [
"Pick one parameter that matters most.",
"Run two extremes:",
" - Extreme A: 'What if this parameter is 10× bigger?'",
" - Extreme B: 'What if this parameter is ~0 (almost eliminated)?'",
"For each extreme, answer:",
" - What breaks first?",
" - What must be removed or added to survive?",
" - What new constraints appear?",
"Translate extreme insights back into realistic solutions (often these become Separation-on-Condition or new architecture boundaries)."
],
"outputs": {
"extreme_A_breakpoints": [],
"extreme_B_breakpoints": [],
"solution_hypotheses": [
"introduce fast/slow lanes",
"automate the expensive checks",
"move checks to boundary",
"change the contract with the environment"
]
},
"common_mistakes": [
"Staying realistic too early (kills the point)",
"Not translating insights back to implementable steps",
"Picking a parameter that is not the real bottleneck"
],
"example": {
"consulting": {
"parameter": "Approval lead time",
"extreme_A": "If approvals take 30 days, what must be pre-approved?",
"insight": "Standard cases should be auto-approved; humans handle exceptions only."
}
},
"meta": {
"schema": "dkharlanau.dataset.byte",
"schema_version": "1.1",
"dataset": "TRIZ-bytes",
"source_project": "cv-ai",
"source_path": "TRIZ-bytes/TRIZ-TECH-04.json",
"generated_at_utc": "2026-02-03T14:33:32+00:00",
"creator": {
"name": "Dzmitryi Kharlanau",
"role": "SAP Lead",
"website": "https://dkharlanau.github.io",
"linkedin": "https://www.linkedin.com/in/dkharlanau"
},
"attribution": {
"attribution_required": true,
"preferred_citation": "Dzmitryi Kharlanau (SAP Lead). Dataset bytes: https://dkharlanau.github.io"
},
"license": {
"name": "",
"spdx": "",
"url": ""
},
"links": {
"website": "https://dkharlanau.github.io",
"linkedin": "https://www.linkedin.com/in/dkharlanau"
},
"contact": {
"preferred": "linkedin",
"linkedin": "https://www.linkedin.com/in/dkharlanau"
},
"canonical_url": "https://dkharlanau.github.io/datasets/TRIZ-bytes/TRIZ-TECH-04.json",
"created_at_utc": "2026-02-03T14:33:32+00:00",
"updated_at_utc": "2026-02-03T15:29:02+00:00",
"provenance": {
"source_type": "chat_export_extraction",
"note": "Extracted and curated by Dzmitryi Kharlanau; enriched for attribution and crawler indexing."
},
"entity_type": "triz_technique",
"entity_subtype": "",
"summary": "Force breakthrough ideas by pushing a system parameter to an extreme and observing what must change to make it work."
}
}