Cities & Infrastructure · Infrastructure Failure Database
Record failures, near misses, causes, impacts and corrective action so the same preventable breakdown is not repeated.
This page turns the subject into a public problem-solving framework: place, problem, causes, evidence, solutions, roadmap, results and contribution.
Problem
What is failing, for whom and where?
02 · Near misses and recurring defects are rarely analysed together.
03 · Lessons may not change design, procurement, operation or maintenance standards.
Root causes
Test causes before selecting projects.
Blame culture discourages reporting
Verify this mechanism against local institutions, assets, user experience and available evidence.
Incident definitions differ
Verify this mechanism against local institutions, assets, user experience and available evidence.
Contract and design evidence is fragmented
Verify this mechanism against local institutions, assets, user experience and available evidence.
Investigations stop at immediate cause
Verify this mechanism against local institutions, assets, user experience and available evidence.
Corrective actions lack owners and deadlines
Verify this mechanism against local institutions, assets, user experience and available evidence.
Legal concerns prevent useful learning
Verify this mechanism against local institutions, assets, user experience and available evidence.
Evidence
Record what is known, how it is known and what remains uncertain.
| Evidence area | What to establish | Key limitation |
|---|---|---|
| Event | Time, place, asset and failure mode | Initial reports are provisional |
| Impact | People, service, cost and environment | Indirect effects may emerge later |
| Cause | Immediate, contributing and systemic factors | Cause needs competent review |
| History | Prior defects, maintenance and warnings | Missing records are themselves a signal |
| Correction | Action, owner, due date and verification | Recommendation is not completion |
Evidence rule: label verified evidence, partial evidence, illustrative analysis, community submission and insufficient evidence separately.
Solutions
Combine immediate action, controlled pilots and structural reform.
Define failure and near-miss categories
Define owner, cost, dependency, safeguard, baseline and decision gate.
Protect evidence and urgent safety reporting
Define owner, cost, dependency, safeguard, baseline and decision gate.
Pilot independent root-cause reviews
Define owner, cost, dependency, safeguard, baseline and decision gate.
Link failures to assets, contracts and standards
Define owner, cost, dependency, safeguard, baseline and decision gate.
Publish anonymised learning summaries
Define owner, cost, dependency, safeguard, baseline and decision gate.
Track corrective action until verified closure
Define owner, cost, dependency, safeguard, baseline and decision gate.
Roadmap
Move from diagnosis to accountable improvement.
Define place, people and outcome
Record the responsible actor, evidence requirement and next decision.
Build a verified baseline
Record the responsible actor, evidence requirement and next decision.
Diagnose causes and constraints
Record the responsible actor, evidence requirement and next decision.
Compare options and pilot
Record the responsible actor, evidence requirement and next decision.
Deliver with safeguards
Record the responsible actor, evidence requirement and next decision.
Measure, improve and scale
Record the responsible actor, evidence requirement and next decision.
Results
Track outcomes people can experience.
Failures recorded
Publish baseline, target, actual, date, geography, source and distribution.
Near misses reported
Publish baseline, target, actual, date, geography, source and distribution.
Root causes completed
Publish baseline, target, actual, date, geography, source and distribution.
Corrections verified
Publish baseline, target, actual, date, geography, source and distribution.
Repeat failure rate
Publish baseline, target, actual, date, geography, source and distribution.
Learning adopted
Publish baseline, target, actual, date, geography, source and distribution.
Risks & safeguards
Address foreseeable harm before scaling.
Blame and concealment
Reporting triggers punishment rather than learning
Safeguard: Separate learning review from misconduct processPrivacy and liability
Sensitive cases expose people
Safeguard: Redact and control accessPremature conclusion
Early narratives become fixed
Safeguard: Version findings by review stageDatabase without reform
Records grow but standards do not change
Safeguard: Assign policy and engineering ownersFAQ
Questions that should be answered before action.
No. Its purpose is safety learning and accountability with lawful privacy protections.
They reveal system weakness before severe harm occurs.
Only after corrective action is implemented and independently checked where risk warrants.
Improve this Infrastructure Failure Database analysis with local evidence.
Submit a place, source, correction, working practice, implementation lesson or measured result.