From Requirements Management
to Requirements Intelligence

Engineering the World’s Hardest Problems

Quest Global is a global engineering services provider headquartered in Singapore, working at the intersection of deep domain expertise and digital and AI-powered engineering. For more than 25 years, Quest Global has partnered with aerospace and defense, automotive, energy, healthcare, rail, and semiconductor organizations to solve their most complex engineering challenges.

Within aerospace and defense, Quest Global supports programs across systems engineering, avionics, flight controls, safety, certification, and supply chain integration — combining engineering depth with the IntelligenceIQ platform to bring continuous, AI-driven intelligence to programs that were previously managed through static documents and manual review.

Learn more at www.quest-global.com

From Managing Documents to Managing Risk

Aerospace organizations are experiencing unprecedented growth in system complexity. Modern aircraft, rotorcraft, autonomous systems, avionics platforms, and mission systems contain millions of interconnected hardware, software, and systems engineering artifacts that must remain synchronized across decades-long development and operational lifecycles.

Despite heavy investment in requirements management tools, significant challenges remain across nearly every program:

• Inconsistent and ambiguous requirements
• Missing requirements and edge cases
• Poor traceability across the engineering chain
• Certification delays and supplier disconnects
• Costly integration failures and escalating technical debt

The aerospace industry has largely solved document management. It has not solved requirements intelligence.

This paper examines the root causes of requirements failures and proposes a new operating model built on artificial intelligence, engineering ontologies, knowledge graphs, digital threads, and digital coworkers — a model that continuously monitors requirements quality, traceability, compliance, and change impact throughout the engineering lifecycle, shifting programs from reactive document management to proactive risk management.

Requirements Are the Foundation

Requirements serve as the foundation of aerospace engineering. Every design decision, software implementation, verification activity, safety assessment, certification review, and operational procedure ultimately traces back to a requirement.

When requirements are incomplete, ambiguous, inconsistent, or disconnected from downstream engineering activities, the resulting impact can be substantial:

• Schedule delays
• Increased certification effort
• Integration failures
• Supplier misalignment
• Program cost overruns
• Reduced system reliability

Historically, organizations have focused on managing requirements through repositories and document control systems. The next evolution is Requirements Intelligence.

Requirements Intelligence shifts the focus from storing requirements to understanding them, evaluating them, monitoring them, and predicting the consequences of change before risk propagates throughout the program.

The Cost Curve Bends Sharply

Numerous aerospace studies have shown that defects discovered during system integration, verification, and certification often originate from requirements deficiencies rather than implementation defects. Typical failure modes include:

• Missing requirements
• Ambiguous language
• Contradictory requirements
• Missing edge cases
• Incomplete decomposition
• Missing verification criteria
• Broken traceability

The cost of correcting these defects increases sharply across the lifecycle. A requirement defect discovered during concept development may take hours to resolve. The same defect, discovered during flight testing, may require software redesign, hardware modifications, re-verification, supplier updates, and certification rework.

As systems become increasingly software-defined, requirements quality becomes one of the strongest predictors of program success.

Part of a Certification Evidence Chain

Unlike most commercial software products, aerospace systems must satisfy extensive certification and safety requirements, including:

• DO-178C
• DO-254
• ARP4754A
• ARP4761
• MIL-STD-882E
• FAA airworthiness requirements
• Military airworthiness standards

A requirement in an aerospace program is not simply a statement of functionality — it becomes part of a certification evidence chain. For every requirement, certification authorities may ask:

The ability to answer these questions rapidly and accurately directly impacts certification success.

Five Recurring Structural Problems

Across aerospace programs, requirements failures tend to trace back to five recurring structural problems.

Problem 1: Ambiguity

Consider the requirement: “The aircraft shall provide obstacle avoidance capability.” While seemingly clear, multiple interpretations are possible. What constitutes an obstacle? At what range must it be detected? Under what environmental conditions? What sensors are involved? What confidence level is required? Different engineering teams often interpret the same requirement differently, producing inconsistent implementation and verification.

Problem 2: Missing Logic

Requirements often focus on normal operating conditions. However, aerospace systems must also account for abnormal and degraded conditions — sensor failures, GPS degradation, communication loss, operator errors, simultaneous failures, and unexpected environmental conditions. These scenarios frequently emerge during integration testing, when correction costs are highest.

Problem 3: Broken Traceability

Modern aerospace programs contain relationships among customer requirements, system requirements, subsystem requirements, software requirements, hardware requirements, architecture models, source code, test procedures, verification evidence, and certification artifacts. When requirements change, understanding downstream impacts becomes increasingly difficult, and many organizations still rely on manual analysis.

Problem 4: Supply Chain Fragmentation

Large aerospace programs involve hundreds of suppliers, and requirements flow through multiple organizations. Each supplier interprets requirements, decomposes them, adds derived requirements, and implements unique verification approaches. Over time, intent drift occurs, and the final implementation may no longer align with the original system intent.

Problem 5: Certification Evidence Gaps

Certification depends on complete evidence chains. Missing links between requirements, design, verification, and validation can trigger extensive manual audits and schedule delays — and organizations often discover these gaps late in the certification process.

From Storage to Risk Identification

Organizations have invested heavily in tools such as IBM DOORS, DOORS Next, Cameo, Teamcenter, Windchill, Jira, Confluence, Git repositories, and test management systems. These platforms successfully store information. However, they do not actively identify engineering risk.

The key question has shifted:

Five Capabilities, One Discipline

Requirements Intelligence combines artificial intelligence, engineering ontologies, knowledge graphs, digital threads, and digital coworkers to continuously evaluate engineering artifacts and identify emerging risks.

The objective is not to replace engineers. The objective is to augment engineering teams with continuous intelligence. Instead of managing documents, organizations manage risk.

One Concept, Many Vocabularies

One of the largest barriers to engineering consistency is terminology. Different teams often describe identical concepts using different language — for example, “Sense and Avoid,” “Obstacle Avoidance,” “Hazard Detection,” and “Collision Prevention” may all refer to the same underlying engineering concept. Traditional keyword search struggles to connect them.

Engineering ontologies establish semantic relationships among systems, functions, components, requirements, failure modes, test procedures, and certification artifacts. The result is a common engineering language across the enterprise.

Four Coworkers, Continuous Coverage

Digital coworkers continuously monitor engineering ecosystems and identify issues before they become program risks.

Requirements Quality Coworker

Identifies ambiguous language, missing acceptance criteria, non-testable requirements, duplicate requirements, and incomplete requirements.

Traceability Coworker

Monitors missing links, orphan requirements, verification gaps, and incomplete decomposition.

Certification Coworker

Continuously evaluates DO-178C and DO-254 readiness, ARP4754A alignment, and evidence completeness.

Change Impact Coworker

Evaluates impacted systems, software, hardware, tests, and certification artifacts whenever a requirement changes.

Requirements Intelligence for a Next-Generation Autonomous Rotorcraft Program

Program Overview

Consider a next-generation rotorcraft platform supporting:

• Autonomous flight
• Optionally piloted operations
• Military missions
• Commercial operations
• Urban air mobility
• Degraded visual environments

The program involves systems engineering, software engineering, flight controls, avionics, safety engineering, cybersecurity, verification teams, certification teams, and multiple Tier 1 and Tier 2 suppliers. The challenge is maintaining alignment across millions of engineering artifacts.

Mission Requirement

A customer requirement states:

“The aircraft shall autonomously detect and avoid obstacles during low-altitude operations.”

At first glance the requirement appears complete. However, important questions remain unanswered: What qualifies as an obstacle? What environmental conditions apply? What response time is acceptable? What happens during sensor disagreement? What confidence threshold is required? Requirements Intelligence automatically identifies these missing attributes.

System Requirement Decomposition

The mission requirement is decomposed into system-level requirements, for example:

“The aircraft shall detect stationary obstacles greater than one meter in height at a range of 500 meters with 95% confidence.”

“The aircraft shall initiate avoidance maneuvers within 500 milliseconds of obstacle classification.”

“The aircraft shall maintain obstacle avoidance capability following any single sensor failure.”

Digital coworkers automatically evaluate these requirements for clarity, completeness, testability, and compliance before formal reviews occur.

Ontology-Based Engineering Context

Requirements become linked to systems (Flight Controls, Navigation, Mission Computer, Sensor Fusion), components (Radar, LiDAR, EO/IR Sensors, Flight Computers), failure modes (Sensor Loss, False Positives, False Negatives, GPS Failure), and certification domains (Safety, Software, Hardware, Human Factors). The requirement becomes part of a living engineering knowledge graph.

Software Development

Software teams derive software requirements from system requirements. Digital coworkers verify consistency, traceability, requirement completeness, and alignment with parent requirements — identifying potential gaps before implementation begins.

Verification Planning

Verification teams develop simulation tests, hardware-in-the-loop tests, integration tests, and flight tests. Digital coworkers identify requirements lacking tests, tests lacking requirements, missing evidence, and duplicate verification activities, so certification readiness improves continuously.

Supplier Integration

Requirements flow to multiple suppliers. Digital coworkers continuously monitor supplier requirements against the program’s baseline requirements, identifying missing derived requirements, contradictory interpretations, missing evidence, and terminology drift. Potential integration failures are identified months before formal integration testing.

Managing Change

A new customer requirement is introduced:

“The aircraft shall detect power transmission lines during low-altitude operations.”

Traditionally, this impact analysis might require weeks. The Change Impact Coworker performs the analysis automatically — within minutes identifying every downstream consequence across systems, software, suppliers, tests, and certification artifacts.

Engineering teams gain immediate visibility into downstream consequences — turning a multi-week manual exercise into a continuous, automated capability.

Proactive Alerts, Not Dashboards

Most organizations rely on dashboards. Engineers rarely have time to monitor them continuously. The future operating model is different.

Digital coworkers continuously monitor requirements repositories, MBSE models, Jira defects, source code, test systems, supplier portals, and certification evidence. Rather than engineers searching for information, the system proactively surfaces it.

Instead of searching for information, engineers receive prioritized recommendations — ranked by risk, tied to a category, and ready for action.

An Intelligence Layer Above Existing Systems

The Requirements Intelligence Platform operates as a unified intelligence layer above existing engineering systems — it does not replace DOORS, Cameo, Teamcenter, or Jira; it sits above them, continuously reading and reasoning across them.

At the foundation, existing engineering systems — DOORS, DOORS Next, Cameo, Teamcenter, Windchill, Jira, Git, test systems, and supplier portals — continue to operate unchanged. Above them, the intelligence layer applies engineering ontologies, aerospace knowledge graphs, digital threads, large language models, and retrieval systems to make sense of that data. Digital coworkers — covering requirements quality, traceability, certification, change impact, and supplier compliance — turn that understanding into continuous monitoring. The Aerospace Engineering Control Center sits at the top, providing alerts, recommendations, risk prioritization, compliance monitoring, and executive visibility.

Measurable Improvement, Four Ways

Organizations implementing Requirements Intelligence can expect improvements across four dimensions.

Quality

Improved requirements consistency, reduced ambiguity, and improved completeness.

Program Execution

Faster impact analysis, reduced integration failures, and reduced engineering rework.

Certification

Improved audit readiness, faster compliance assessments, and reduced certification risk.

Knowledge Retention

Preservation of institutional knowledge, reduced dependency on tribal expertise, and improved onboarding of engineering teams.

Better Intelligence, Not More Documentation

For decades, the aerospace industry has focused on managing requirements. The next generation of engineering organizations will focus on understanding them.

By combining AI, engineering ontologies, digital threads, knowledge graphs, and digital coworkers, organizations can transition from reactive, document-centric engineering to proactive, intelligence-driven engineering. The result is:

• Better requirements
• Stronger traceability
• Faster certification
• Reduced program risk
• Improved supplier alignment
• Accelerated delivery of next-generation aerospace systems

The future of aerospace engineering is not more documentation. It is better intelligence.