Write & Check Requirements

Write & Check Requirements Overview

Saphira’s Write & Check Requirements feature provides a comprehensive workflow for refining requirements through classification, quality checking, traceability linking, and export. This workflow is used both for manually written requirements and for refining requirements extracted from documents.

Getting Started

From the Home Page, navigate to the Seek Guidance section and click Write & Check Requirements, or use this workflow after extracting requirements from documents.

Write & Check Workflow

The Write & Check Requirements workflow guides you through a structured process:

Input → Classify → INCOSE Lint → Trace → Conflicts → Export

Input

Enter your requirement text manually, or import from:

Extracted requirements from documents
CSV files with requirement tables
Jama Connect integration
Manual entry

Classify

AI automatically suggests requirement classification level:

Stakeholder
System
Subsystem
Component
Hardware/Software

Review the suggested classification and rationale, and adjust if needed.

INCOSE Lint

Quality check against INCOSE requirements engineering principles:

Atomicity: Single capability per requirement
Verifiability: Testable with single procedure
Clarity: No ambiguous terms
Suggested rewrites for improvement

Review lint results and apply suggested corrections.

Trace

AI suggests trace links to:

Related safety analyses (FMEA failure modes, HARA hazards)
Existing requirements at adjacent levels
Test cases and verification activities
Standards clauses

Select relevant traces to establish traceability.

Conflicts (CSV imports only)

For CSV imports, check for conflicts with existing requirements:

Duplicate requirement detection
Contradictory requirements identification
Overlapping scope warnings

Export

Import validated requirements to your project database:

Full traceability to source document
Version control
Change tracking
Integration with Traceability Matrix

Classification

Automatic Classification

Saphira automatically analyzes requirement text to suggest the appropriate classification level: Requirements Classification View

Classification Analysis

The system examines requirement text for:

User Perspective Terms: user, operator, driver, passenger, customer
Technical Terms: voltage, current, frequency, bandwidth, latency
System-Level Scope: system shall, the equipment, overall performance
Component Terms: ECU, sensor, actuator, module, controller
Subsystem Indicators: interface, communication, integration, connection
Hardware Terms: physical, electrical, mechanical, thermal, environmental
Software Terms: algorithm, function, code, memory, processing

Dynamic Recommendations

Based on analysis, Saphira provides recommendations:For Stakeholder Level:

Add user perspective terms
Remove technical implementation details
Broaden scope to system-level concerns

For System Level:

Add system-level scope indicators
Remove component-specific details
Add technical terms for specificity

For Component Level:

Add component-specific terms
Narrow scope to component concerns
Include measurable parameters

INCOSE Linting

Quality Checking

Saphira provides automated requirement quality checking against INCOSE best practices: INCOSE Linting View

INCOSE Principles

Requirements are checked against:

Atomicity: One capability per requirement
Verifiability: Testable with single procedure
Clarity: No ambiguous terms (avoid “appropriate”, “adequate”)
Single Safe State: One safe state per requirement
Shall Statements: Active voice with “shall”

Suggested Rewrites

When issues are detected:

Specific suggestions for improvement
Before/after comparisons
Rationale for each suggestion
One-click apply to update requirement text

Requirement Quality Rules

Principle	Description	Example
Atomicity	One capability per requirement	✅ “System shall detect obstacles”
Verifiability	Testable with single procedure	✅ “Response time shall be ≤100ms”
Clarity	No ambiguous terms	❌ Avoid “appropriate”, “adequate”
Single Safe State	One safe state per requirement	✅ “System shall enter safe state A”
Shall Statements	Active voice with “shall”	✅ “The system shall provide…”

Traceability

Automatic Trace Suggestions

When creating requirements, Saphira suggests traces to:

Safety Analyses: FMEA failure modes, HARA hazards, TARA assets
Existing Requirements: Requirements at adjacent levels in the hierarchy
Test Cases: Verification activities and test procedures
Standards Clauses: Relevant standards and compliance requirements

Traceability Matrix Integration

After saving requirements, view them in the Traceability Matrix:

Parent-child decomposition trees
Cross-level traceability
Verification coverage analysis
Gap identification for missing traces

Import Sources

Manual Entry

Type or paste requirement text directly into the input field. Best for:

Single requirements
Quick requirement entry
Requirements from external sources

CSV Import

Import requirements from CSV files with:

Requirement IDs and descriptions
Parent-child relationships
Classification levels
Additional metadata

The workflow includes conflict checking for CSV imports to detect duplicates and contradictions.

Jama Connect

Import requirements directly from Jama Connect:

Automatic synchronization
Requirement metadata preservation
Traceability link maintenance

Extracted Requirements

After extracting requirements from PDFs or images, use Write & Check to:

Classify extracted requirements
Validate quality with INCOSE linting
Establish traceability links
Import to project database

Post-Export Workflow

After exporting requirements, you can:

View in Traceability Matrix: See your requirements in the full traceability context
Extract More Requirements: Continue extracting from additional documents
Run Safety Analysis: Use requirements for FMEA, HARA, or TARA analysis
Generate V&V Plan: Create verification and validation plans
Build Safety Case: Create GSN safety case structures

Extracted Requirements Export Options View

Integration with Safety Workflows

Refined requirements integrate directly with Saphira’s safety analysis:

FMEA: Requirements become focus elements for failure mode analysis
HARA: Requirements populate hazard sources and control measures
TARA: Interface definitions feed cybersecurity asset identification
Gap Analysis: Extracted specifications enable automated standards compliance checking
Safety Case: Requirements become evidence in GSN safety arguments

Extract Requirements

Extract requirements from PDF documents and images using AI

Requirement Refinement

Decompose requirements hierarchically from stakeholder to SW/HW level

Import from Jama

Pull requirements from Jama, clean with AI, and sync back

Traceability Matrix

View and manage requirement relationships and traceability

Get Started

Extraction & Requirements

Safety Analysis

Verification & Compliance

Safety Case & Standards

Write & Check Requirements