Safety and risk analysis
Safety cases built on live engineering evidence
Argue why your system is acceptably safe with a claims, arguments and evidence structure. Evidence links to the records in your other analyses instead of copies, so the case shows when something it relies on has changed.
Overview
A safety case is a structured argument, supported by evidence, that a system is acceptably safe for a given use in a given environment. A top claim is broken down into sub-claims, each supported by arguments and, finally, by evidence such as analyses, test results and requirements.
In ModelGraphX each case has a defined scope, boundaries, operating context, lifecycle phase, objective, acceptance basis and acceptance authority. The claim tree holds claims, arguments, assumptions, defeaters and uncertainties, and evidence points to live Risk Registry entries, requirements, hazards, FMEA rows and bowtie elements.
When linked sources change, change-impact recompute shows which parts of the argument are affected. A health label summarises completeness and evidence coverage; it is a working indicator, not a safety certificate. Baselines freeze the case for review and publication.
Capabilities
Scoped case record
Scope, boundaries, operating context, lifecycle phase, objective, acceptance basis and acceptance authority.
Claim tree
Top claim, sub-claims, arguments, assumptions, defeaters and uncertainties in one structure.
Linked evidence
Reference Risk Registry entries, requirements, hazards, FMEA rows and bowtie elements rather than copying them.
Argument patterns
Reuse platform patterns, including an ALARP and risk-reduction pattern, and record acceptance criteria.
Starter templates
Barrier adequacy, residual risk acceptance, safety requirements satisfaction, hazard identification completeness and a hydropower dam facility case.
Health label
See completeness and evidence coverage at a glance.
Change impact
Recompute which claims are affected when the evidence they rely on changes.
Baselines and governance
Create and publish baselines and run reviews.
Requirements coverage
Check which requirements the case covers, keep an assurance register and associate assets.
Regulatory reports
Export HTML and Word reports with a claims, arguments and evidence profile, including jurisdiction templates such as the Swedish helhetsbedömning.
Report tab
Export Word, PDF or PowerPoint reports with argument figures.
AI assistance
- A preview AI agent offers gap analysis, a draft case structure, argument challenges, an executive narrative and pattern suggestions; you accept or reject each finding.
- Draft report sections and a summary for you to review.
AI proposals are drafts: you review and accept them before they enter your model.
Standards and methods
- Claims, arguments and evidence (CAE)
- ALARP
ModelGraphX supports work done under these standards and methods; it does not certify compliance.
Who uses it
- System safety and assurance engineers
- Dam owners and hydropower operators preparing safety assessments
- Rail, defense and aerospace programmes
- Regulatory and independent safety assessors
- Safety managers at operators of complex facilities
Works with
- Risk Registry
Risk Registry entries serve as evidence for risk acceptance claims.
- Requirements
Requirements coverage shows which requirements the argument relies on.
- Hazard Management
Hazards and studies support hazard identification claims.
- FMEA
FMEA rows can be cited as evidence.
- Bowtie Analysis
Threats, consequences and barriers can be cited as evidence.
- Reports
Safety case content can be part of composed study documents.
Frequently asked questions
- What is a safety case?
- A safety case is a structured argument, supported by evidence, that a system is acceptably safe for a specific application in a specific environment. It typically breaks a top claim into sub-claims with arguments and evidence.
- Does ModelGraphX support GSN?
- ModelGraphX uses a claims, arguments and evidence structure with a GSN-style tree, but it is not a full GSN notation editor.
- What happens when the evidence changes?
- Evidence is linked, not copied. When a linked risk, requirement or hazard changes, change-impact recompute shows which claims are affected, so you can update the argument before the next baseline.
- Does the health label certify my system?
- No. It summarises completeness and evidence coverage to help the team. Acceptance remains a decision for the people and authorities named in the case.
Try it on a real project
Explorer is free, and new accounts get a 30-day trial of the paid capabilities.