AI Leave Aviation. 78% Zero-Touch. Multiple Bargaining Units. One Agent.
Cockpit. Cabin. Ground. EASA FTL. Medical Grounding.
The AI Agent classifies leave type, bargaining unit, and operational status. It distinguishes regulatory mandatory rest from leave, detects Medical Grounding, and checks crew pairing constraints. Calculations run through deterministic rule engines. The human stays where employment law, employee representation, or regulatory requirements demand it.
Bargaining units
Zero-touch rate
FTL compliance
Highest rule complexity
Multiple collective agreements (cockpit, cabin, ground). Zero-touch: Gosign simulation model.
What the Agent classifies
Five dimensions, one Agent
Multiple bargaining units with separate collective agreements, regulatory mandatory rest periods, Medical Grounding, crew pairing as operational constraint, and seasonal blackout periods. You know the complexity. Here is how the AI Agent resolves it:
Multiple collective agreements in parallel
Cockpit, cabin, and ground staff - each bargaining unit has its own collective agreement with separate leave rules, special leave, and compensatory time off. The AI Agent identifies the staff group and applies the correct rule set. No confusion between pilot leave and ground staff leave.
EASA FTL mandatory rest periods
Flight Time Limitations are regulatory mandatory rest periods that must not count as leave. The AI Agent separates cleanly: FTL rest periods run in separate tracking, leave entitlements are not reduced by mandatory rest. During leave planning, the agent checks FTL compliance against the flight schedule. (UK: UK CAA regulations post-Brexit apply separate FTL frameworks for UK-registered operators.)
Medical Grounding
Loss of Medical Certificate (Class 1 cockpit, Class 2 cabin) is not regular sickness but a regulatory absence. The AI Agent classifies Medical Grounding as a separate leave type with collective-agreement-specific consequences. Health data remains architecturally separated - the agent sees only the status, not the diagnosis.
Crew pairing as constraint
Cockpit and cabin must be staffed together for every flight. The AI Agent checks every leave request against crew availability and the flight schedule, including minimum staffing per position. On potential breach: automatic escalation with an alternative suggestion and context for the scheduler.
Seasonal blackout periods
Summer programme and Christmas peak operations mean blackout periods for flight crew. The AI Agent knows the blackout periods from the company governance framework and checks every request against them. Exceptions require a line manager decision - the agent documents the exception in the audit trail.
One leave request in aviation. 10 decision steps.
A pilot submits a leave request. The Leave Decision Layer breaks the transaction into individual decision steps:
| Step | Decision | Decision maker | Rationale |
|---|---|---|---|
| 1 | Identify staff group and collective agreement | AI Agent | Agent identifies: cockpit, cabin, or ground from employee master data |
| 2 | Classify leave type | AI Agent | Agent identifies: annual leave, special leave, FTL rest period, or Medical Grounding |
| 3 | Calculate leave entitlement | Rule engine | Collective-agreement-specific base leave + additional leave (long-haul, seniority) |
| 4 | Check EASA FTL compliance | Rule engine | Minimum rest periods between duties, flight hour accumulation, duty period limits |
| 5 | Check crew pairing and minimum staffing | AI Agent | Agent checks crew availability against flight schedule and minimum staffing per position |
| 6 | Check blackout periods | Rule engine | Summer programme, Christmas peak, company-agreed blackout periods |
| 7 | Generate approval recommendation | AI recommends, human decides | All rules met: recommend approval. Conflict: escalation with context |
| 8 | Check return-to-work threshold | Rule engine | 42+ sick days in 12 months rolling. Medical Grounding tracked as special case separately |
| 9 | Calculate sick pay continuation | Rule engine | Statutory sick pay period per case, collective-agreement-specific top-up where applicable |
| 10 | Generate audit entry | Rule engine | Complete decision record: leave type, collective agreement, FTL status, crew check, result |
Simulation
Calculated for aviation volumes
We configured the Leave Decision Layer with realistic aviation parameters and ran the simulation. Multiple bargaining units, EASA FTL, Medical Grounding, crew pairing.
Simulation parameters
| Employees | 5,000 to 80,000+ (cockpit, cabin, ground) |
| Collective agreements | Separate agreements for cockpit, cabin, and ground staff |
| Regulatory framework | EASA FTL (EU-OPS), Medical Certificates Class 1 and Class 2 |
| Leave types | Annual leave, FTL rest period, Medical Grounding, standby, special leave |
| Constraints | Crew pairing, seasonal blackouts, bases (hubs and outstations) |
| Return-to-work rate | Industry average sickness rate for aviation, Medical Grounding separate |
Before / After
| Dimension | Manual | Decision Layer |
|---|---|---|
| Leave request processing time | 2-5 days (crew check) | < 2 minutes |
| Zero-touch rate | 0% | 78% |
| FTL conflicts detected | Manual by scheduler | Automatic in real time |
| Medical Grounding correctly classified | Error-prone | 100% correct (dedicated leave type) |
| Return-to-work deadline failures | 12-18% | 0% (automatic trigger) |
| Audit readiness | Manually reconstructed | Automatically generated |
EASA FTL: EU Regulation 965/2012 Subpart FTL. Sickness rates: industry averages. Simulation results: Gosign model calculation.
In our simulation, the Decision Layer achieves a zero-touch rate of 78%. The lower rate reflects the high regulatory density: crew pairing conflicts, Medical Grounding classification, and FTL edge cases require human decisions more frequently. For the 78%, a complete, audit-ready decision record is available.
Architecture and implementation
The Leave Decision Layer runs entirely within your infrastructure. For aviation, this means: integration with crew management systems, FTL tracking, medical database, and a complete audit trail. Typical pilot projects start with one staff group (e.g. ground) and expand to cockpit and cabin.
Deep Dive in the Agent Briefing (Gosign Magazine)
Our expert article series for decision-makers deploying AI Agents in the enterprise.
Leave Decision Layer in Other Industries
Every industry has its own collective agreements, its own leave types, and its own complexity drivers. The Decision Layer is the same. The configuration is industry-specific.
Chemicals
Site-level agreements, collective-agreement time-off option, shift-supplement leave, hazardous materials return-to-work
Financial Services
Regulatory mandatory leave, compliance-mandated leave blocks, banking holidays, phased retirement
Retail
6-day working week, 64% part-time ratio, seasonal blackout periods, variable-hours contracts
Frequently Asked Questions about the Aviation Configuration
How does the Decision Layer distinguish between multiple bargaining units?
Each employee is assigned to a staff group: cockpit, cabin, or ground. The AI Agent identifies the assignment from the employee master record and applies the correct collective agreement with the respective leave rules. Multiple bargaining units, multiple rule sets, one Agent.
How does the Agent handle EASA FTL rest periods?
EASA Flight Time Limitations are regulatory mandatory rest periods, not leave. The AI Agent distinguishes precisely: FTL rest periods are tracked separately and considered during leave planning. No leave day may replace a regulatory mandatory rest period. FTL compliance is checked in real time against the flight schedule. (UK: UK CAA regulations post-Brexit apply separate FTL rules for UK-registered operators.)
What happens during Medical Grounding?
When a pilot or cabin crew member loses their Medical Certificate, the AI Agent classifies the absence correctly: Medical Grounding is not regular sickness but a regulatory restriction. The rule engine calculates the employment law consequences (sick pay continuation, operational fitness) per collective agreement. Health data remains architecturally separated.
How does crew pairing work with leave planning?
Crew pairing is the operational constraint: cockpit and cabin must be staffed together, every position has minimum staffing. The AI Agent checks every leave request against crew availability and the flight schedule. On potential breach: escalation with context, not blanket rejection.
How are seasonal blackout periods handled?
Seasonal blackout periods (summer programme, Christmas peak) are stored as company governance frameworks. The AI Agent knows the blackout periods and checks every request against them. Exceptions are possible but require a line manager decision. The agent documents the exception in the audit trail.
Is the system compatible with employee representation requirements?
Yes. In aviation, separate employee representation bodies often exist per bargaining unit. The Leave Decision Layer makes the rule engine of each collective agreement transparent, rejection reasons traceable, and reports pseudonymised. Co-determination (Mitbestimmung) rights over leave policies are fully supported. (US: No federal co-determination requirement. FMLA provides 12 weeks unpaid leave as baseline.)
Let us run the numbers.
30 minutes. Your collective agreements, your crew constraints, your result. We configure the Decision Layer with your actual parameters.
Book a call