Documentation

System Overview

S-System is an AI-powered platform for airline operations management designed to optimize passenger experiences, maximize operational efficiency, and enhance decision-making through advanced analytics and predictive intelligence.

Seven models run against your operation. They forecast disruption and delay up to four days (96 hours) before departure, project passenger flow and terminal congestion, anticipate demand by route, group passengers into segments worth treating differently, and choose which ancillary offer to put in front of each one. Seven engines work alongside them: they follow each aircraft’s rotation, size crew reserves and watch crew fatigue, test fuel tankering on every leg, estimate no-shows and the overbooking level each flight can carry, measure the belly space left for cargo, and read how demand responds to fares by route.

All of it runs on the systems you already have, fed by live operational and market data rather than a static schedule.

Built on a modular architecture, S-System offers scalable solutions for airlines of all sizes, from regional carriers to major international airlines.

Real-time flight tracking and delay prediction
Passenger segmentation and personalization
Airport operations optimization, with sensor and survey readings where the airport shares them
Advanced analytics and reporting dashboards
Disruption management and rebooking automation
Multi-tenant company and user management

Key Features

Flight Operations Dashboard

Real-time flight tracking with predictive delay estimation and gate optimization.

• Comprehensive flight status monitoring
• AI-powered delay prediction
• Automated gate assignment optimization
• Weather impact assessment

Passenger Analytics

Detailed passenger segmentation and personalization capabilities.

• Customer Intelligence segmentation
• Personalized service recommendations
• Upgrade opportunity identification
• Loyalty-tier recognition in every offer, with retention analysis once a loyalty system is connected

Airport Operations

Airport monitoring and optimization from your operational data, and from the sensor and survey readings an airport shares.

• Real-time passenger flow tracking
• Congestion prediction and resolution
• Retail engagement analytics
• Dynamic staffing recommendations

Predictive Analytics

Advanced forecasting and prediction engines for operations.

• Route profitability analysis
• Delay hotspot identification
• Demand forecasting engine
• Custom predictive models

Business Intelligence

Comprehensive reports and analytics dashboards.

• Operational performance reports
• Financial analysis dashboards
• Custom report builder
• Export to CSV/PDF formats

Dynamic Offer Pricing

Ancillary revenue priced to the moment the passenger is in.

• Offers priced from delay, timing and load
• Loyalty recognition built in
• Every price movement explained
• Conversion tracked against each offer

Disruption Management

Proactive disruption prediction and automated response.

• Disruption Prediction Engine
• Automated rebooking suggestions
• Impact assessment scoring
• Recovery optimization

Predictive Intelligence Suite

S-System's Predictive Intelligence Suite uses advanced AI to deliver actionable insights across your airline operations. Our proprietary engines are trained on comprehensive data from multiple integrated APIs to provide accurate, real-time predictions.

Predictive intelligence

How S-System turns your operation into forecasts

Forecasts built from your own operation and checked against what actually happened.

Connect

Works with what you run

Your operations systemsFlights, aircraft and schedule changes

Your reservation systemBookings and passengers on each flight

Live operational contextConditions at and around your network

Read-only by default. Nothing in your systems changes without your approval.

Prepare

Decision-ready data

Validated automaticallyGaps and inconsistencies caught early

One consistent pictureEvery system and station on the same footing

Signals that matterYour network's patterns, learned from your history

Quality checks run before anything reaches a forecast.

Predict

Seven models, seven engines

AI models

Disruption risk

Delay forecast

Passenger flow

Congestion

Demand

Segmentation

Ancillary offers

Operational engines

Rotation risk

Crew reserve

Crew fatigue

Fuel tankering

Commercial engines

No-shows and overbooking

Cargo capacity

Fare elasticity

Forecasts up to four days (96 hours) before departure.

Act

Insight where it is used

Alerts for your teamsFor the flights and stations each team covers

Passenger rebookingAlternatives found and sent once approved

Compensation exposurePassenger rights cost, priced per flight

Reports and APIBoard-ready reports and a documented API

Every change to your systems is approved by your people.

Measure

Accuracy you can verify

Every forecast checkedAgainst what actually happened

Accuracy on your dataOpen to your administrators at any time

Continuous improvementForecasts sharpen as your operation evolves

Performance reported openly, on your own operation.

96hForecast horizon

7AI models, plus seven operational and commercial engines

9Passenger rights regimes priced

4,920Airports worldwide

The prediction suite

Disruption risk Delay forecast Passenger flow Congestion Demand Segmentation Ancillary offers Rotation risk Crew reserve Crew fatigue Fuel tankering No-shows and overbooking Cargo capacity Fare elasticity

Disruption Prediction Engine

Anticipate operational disruptions before they impact your passengers and operations.

Weather impact analysis
Traffic pattern recognition
Airport congestion monitoring
Crew availability tracking

Delay Prediction Engine

Forecast flight delays with precision to enable proactive passenger communication.

Real-time ATC status integration
Historical delay pattern analysis
Weather correlation mapping
Ground operations tracking

Customer Intelligence Engine

Segment passengers intelligently to deliver personalized experiences at scale.

Behavioral segmentation
Value-based clustering
Preference prediction
Loyalty tier optimization

Passenger Flow Engine

Optimize terminal operations with real-time passenger movement insights.

Sensor and survey readings, where the airport shares them
Queue time prediction
Heatmap visualization
Zone capacity management

Demand Forecasting Engine

Predict passenger demand to optimize capacity planning and pricing strategies.

Seasonal trend analysis
Event impact modeling
Holiday traffic prediction
Route demand optimization

Revenue Optimization Engine

Maximize ancillary revenue with personalized product recommendations.

Personalized upsell suggestions
Ancillary propensity scoring
Dynamic ancillary pricing, alongside your revenue management
Cross-sell optimization

Commercial Forecasting Engine

The commercial side of every flight, forecast from the bookings and operations data you already send.

No-shows per flight, with the overbooking level priced against denied-boarding cost
Belly space left for cargo after passengers and bags
Fare elasticity by route, booking window and season
Route and market level, with no passenger identity in any forecast

Real-Time Data Integration

Our Predictive Intelligence engines integrate with multiple real-time data APIs to ensure accuracy and relevance:

Weather Services

Real-time weather APIs

Flight Data

Aviation data providers

ATC Status

Air traffic control feeds

Events & Holidays

Event data services

System Architecture

Access

Wherever your teams work

Web applicationOn desktop and phone

REST APIFor the systems you run

AlertsTo the right people, on time

ReportsPDF and CSV, any period

Operations modules

One view of the whole operation

Flight operationsStatus, risk and rotations

Disruptions and rebookingFrom alert to recovery

Airport operationsStations, crew and fuel

Passenger analyticsSegments and personalization

AnalysisPerformance and revenue

AdministrationUsers, connections, settings

Predictive intelligence

Forecasts that improve over time

AI models and enginesSeven models and seven engines

Continuous learningSharpens as your data grows

Accuracy trackingEvery forecast checked

Quality controlChecked before going live

Integration

Connected to your systems

Operations systemsFlights and schedules

Reservation systemsBookings and passengers

Live flight dataStatus across the network

Approved updatesSent back after sign-off

Your data

Private and portable

IsolatedVisible to your airline only

EncryptedProtected at rest

RedundantReplicated and backed up

PortableA full copy, whenever you want

S-System Architecture Overview

S-System is built in five layers, with governance controls that apply across all of them.

Access

Everything your teams touch: dashboards that work on a phone at the gate as well as a desk, a documented API for the systems you already run, timely alerts, and reports in PDF or CSV.

Operations modules

Flight operations, disruptions and rebooking, airport operations, passenger analytics and analysis, all in one place and all built on your airline's own data.

Predictive intelligence

Seven AI models and seven engines, four operational and three commercial, across your operation, improving continuously as your data grows. Every forecast is checked against what actually happened, and the results are open to your administrators.

Integration

S-System connects to the operations and reservation systems you already run. Updates go back to your systems only after your own people approve them.

Your Data

Your data is held encrypted and replicated, isolated from every other carrier on the platform, and backed up continuously. Retention follows your policy, and you can take a full copy with you whenever you want it.

Compliance and Assurance

Disruption is a regulatory event as much as an operational one. S-System prices statutory exposure per flight under the regime that actually governs it, across nine jurisdictions, and backs its forecasts with measured performance you can inspect.

Passenger rights, nine jurisdictions

A delayed flight from Frankfurt carries EUR 600 per passenger under EU261. The same aircraft on a domestic US sector carries no statutory compensation at all. S-System resolves which regime governs each flight from its route and operating carrier, across 4,920 airports.

EU261 UK261 Canada APPR Turkey SHY Saudi GACA US DOT India DGCA Brazil ANAC UAE GCAA

Published prediction accuracy

Every prediction is scored against the recorded outcome and the result is yours to see: how often a warning was right, how many disruptions were caught in advance, and how much notice you had, broken down by airport and by route.

Most forecasting tools ask you to take their accuracy on trust. S-System hands you the evidence, so you can point operational decisions at the routes and lead times where the model is demonstrably strong, and hold us to it.

Audit trail

Every privileged action is recorded: who did it, when, and what it affected. Access to passenger data is held to a higher standard again, so if a regulator or your own compliance team asks who saw a record and when, the answer is already there. Retention is configurable to match your obligations.

Your data stays yours

Your operational data is yours, and it is isolated from every other carrier on the platform by design rather than by policy. Data is encrypted at rest and in transit, with redundant storage so nothing is lost to a single failure. You can take your data with you at any time.

Dynamic Offer Pricing

A lounge pass is worth more to a passenger facing a four hour delay than to one departing on time. S-System prices ancillary offers from the situation the passenger is actually in, and explains every movement.

Priced from the moment

Offers respond to delay length, time to departure, load factor and route type. A meal voucher during a long wait, a fast-track pass when the connection is tight, an upgrade when the cabin has space.

Every quote carries the reasons behind it, so an offer can explain itself to the passenger and to you.

We price the offer, not the passenger

Loyalty recognition moves prices down. Nothing moves them up because of who is buying. There is no willingness-to-pay model charging your best customers more for the same product.

You control it. Personalized pricing is off until you turn it on, from a single switch in your company settings, and you can turn it off again at any time. While it is on, every affected quote is flagged so your booking flow can carry the disclosure the EU and other markets require.

Revenue you can defend

Each offer is recorded with the price shown, the reasons for it and whether it converted. That turns ancillary pricing from a guess into something you can measure, argue for, and improve.

API Documentation

S-System exposes a documented HTTP API so the systems you already run can read flight status, performance and route analytics directly. Requests and responses are JSON, and every endpoint is scoped to the permissions on your key.

Authentication

API access uses a long-lived key rather than a login exchange. Generate one under Admin → Integrations → API Keys, scoped to the permissions it needs. The key is shown once at creation and stored only as a hash, so it cannot be recovered afterwards.

Send it in the X-API-Key header on every request. Keys carry an expiry, a per-hour rate limit and an explicit permission set, and can be revoked immediately without affecting others.

Request# Every request carries the key
GET /api/v1/flights
X-API-Key: sk_live_your_key_here

# Keys are scoped. A key with only flights:read
# receives 403 on passenger endpoints.

The key is displayed once. It is stored as a hash and cannot be retrieved later. If it is lost, revoke it and generate a replacement.

Response Format

All API responses follow a consistent format:

Response{
  "success": true,
  "data": { ... },
  "message": "Operation successful",
  "quota": { "calls_remaining": 41283, "percent_used": 17.4 }
}

Endpoints for accessing flight data, schedules, and predictions.

Endpoint	Method	Description	Required Plan
`/api/v1/flights`	GET	List flights for an operating day. Filters: `origin`, `destination`, `date` (today by default); up to `limit` rows	Standard
`/api/v1/flights/{flight_id}`	GET	One flight, by its flight number	Standard
`/api/v1/flights/metrics`	GET	On-time performance, average delay and gate conflicts	Standard
`/api/v1/flights/route-analytics`	GET	On-time performance grouped by route	Professional
`/api/v1/flights/search`	GET	Search by flight number or route, in `q`	Standard
`/api/v1/flights/refresh`	POST	Force a refresh of the live flight feed	Professional

What the operations team sees, as data. Dates are YYYY-MM-DD and default to today; times are UTC. Each endpoint is included with the capability shown and called with a key carrying its permission. Every response covers your own airline only.

Endpoint	Method	Parameters	Returns	Requires
`/api/v1/ops/disruptions`	GET	`kind` (daily or sameday), `date`, `station`	The flights most at risk: risk, likely delay, causes, suggested action, recovery options (swaps and retiming, with the delay and cost each saves and the connections it keeps), connections at risk with reprotection, cost including missed connections, and rebooking options	IOCC BriefingKey permission `iocc:read`
`/api/v1/ops/rotation`	GET	`flight` or `tail`, `date`, `origin`, `seed` (a starting delay in minutes, as a what-if)	How a delay travels through the aircraft’s day, leg by leg	Disruption Prediction or IOCC BriefingKey permission `disruptions:read`
`/api/v1/ops/congestion`	GET	`airport` (required), `hours` (up to 48), `start`	Airfield congestion per hour, in UTC	Airport Heatmap or IOCC BriefingKey permission `heatmaps:read`
`/api/v1/ops/passengers`	GET	`date`, `flight`	Booked, connecting and infant passengers per flight. Counts only, never names	Passenger Analytics or IOCC BriefingKey permission `passengers:read`
`/api/v1/ops/crew`	GET	`date`, `station`	Crew duty margin per leg, and reserve crews needed per base. No individual is named	Crew Reserve or IOCC BriefingKey permission `crew:read`
`/api/v1/ops/fuel`	GET	`date`, `flight`	The fuel tankering plan: whether carrying extra fuel pays, with the saving and its CO2 cost	Fuel Tankering or IOCC BriefingKey permission `fuel:read`
`/api/v1/ops/airports`	GET	None	The airline’s stations, busiest first	Airport Heatmap or IOCC BriefingKey permission `airports:read`
`/api/v1/ops/airports/{code}`	GET	`hours` (default 6)	One station: departures at risk in the coming hours, and its congestion	Airport Heatmap or IOCC BriefingKey permission `airports:read`
`/api/v1/ops/forecast`	GET	`hours` (1 to 96, default 24), `station`	Every flight in the window, with its risk and likely delay, forecast for the time remaining to departure, up to 96 hours ahead	IOCC BriefingKey permission `disruptions:read`
`/api/v1/ops/review`	GET	`week` (any day of it), `include=flights`	The weekly review: measures, cost including missed connections, and for a pilot, each target against the week	IOCC BriefingKey permission `iocc:read`
`/api/v1/ops/recovery-decision`	POST	JSON: `date`, `flight_number`, `origin`, `feasible` (true, false or null), `kind`	Record your IOCC’s decision on a flight’s recovery options. S-System learns from each decision which options your IOCC accepts, and ranks future options to match	IOCC BriefingKey permission `iocc:write`
`/api/v1/webhooks/events`	GET	`since`, `type`, `limit`	Signed events from the last 72 hours: list.issued, flight.flagged, review.published, delivery.loaded	Any plan
`/api/v1/webhooks/replay`	POST	JSON: `since` or `event_ids`, optionally `type`, `endpoint_id`	Send events again within 72 hours, signed and marked `X-SSystem-Replay: 1`, with their original ids	Any plan
`/api/v1/ops/capabilities`	GET	None	Every data type the API serves, and whether this airline has it	Any plan
`/api/v1/usage`	GET	None	Calls in the last 30 days, the allowance, and a breakdown by endpoint	Any plan
`/api/v1/subscription`	GET	None	Plan, entitlements and limits for the account	Flight TrackingKey permission `flights:read`

The client libraries cover the flight, airport and usage endpoints; call these directly, as in Try it.

Endpoints for passenger segmentation, insights, and personalization.

Endpoint	Method	Description	Required Plan
Passenger intelligence is delivered through the application and through scheduled data delivery, which suits how most carriers consume it. Talk to your account manager about direct API access for your integration.

Endpoints for airport management, passenger heatmaps and passenger flow analysis.

Endpoint	Method	Description	Required Plan
`/api/airport/{code}/overview`	GET	Get airport overview data	Standard
`/api/airport/{code}/heatmap`	GET	Get the passenger heatmap: measured where the airport shares readings, modelled otherwise	Professional
`/api/airport/{code}/congestion`	GET	Get congestion prediction data	Professional

Every request carries the key. Nothing else is required.

cURL# List flights departing New York
curl -s https://www.s-system.cloud/api/v1/flights?origin=JFK&limit=25 \
  -H "X-API-Key: sk_live_your_key_here"

# On-time performance, average delay and gate conflicts
curl -s https://www.s-system.cloud/api/v1/flights/metrics \
  -H "X-API-Key: sk_live_your_key_here"

# Congestion at a station, with the 1h and 3h forecast (Professional)
curl -s "https://www.s-system.cloud/api/airport/JFK/congestion?hours=24" \
  -H "X-API-Key: sk_live_your_key_here"

Every response is {success, data, message}. Check success before reading data, which is null on an error.

Install the client: pip install kquika-ssystem. The distribution is kquika-ssystem and the import is ssystem.

Pythonfrom ssystem import AuthenticatedClient
from ssystem.api.flight_operations import list_flights, get_flight_metrics

client = AuthenticatedClient(
    base_url="https://www.s-system.cloud",
    token="sk_live_your_key_here",
    auth_header_name="X-API-Key",
    prefix="",            # the key goes in bare, with no "Bearer "
)

with client as c:
    res = list_flights.sync(client=c, origin="JFK", limit=25)

    if not res.success:
        raise SystemExit(res.message)

    for f in res.data.flights:
        print(f.flight_number, f.origin, f.destination, f.scheduled_departure)

    m = get_flight_metrics.sync(client=c).data
    # on_time_percentage is over measured_flights, not total_flights.
    print(f"{m.on_time} on time of {m.measured_flights} measured, "
          f"{m.total_flights} scheduled")

Without the client, any HTTP library works: requests.get(url, headers={"X-API-Key": key}).

Install the client: npm install @kquika-inc/s-system. Node 18 or newer, types included.

JavaScriptimport { client, listFlights, getFlightMetrics } from '@kquika-inc/s-system';

client.setConfig({
  baseUrl: 'https://www.s-system.cloud',
  headers: { 'X-API-Key': process.env.S_SYSTEM_API_KEY },
});

const { data, error } = await listFlights({
  query: { origin: 'LAX', limit: 25 },
});

if (error || !data?.success) {
  throw new Error(data?.message ?? String(error));
}

for (const f of data.data.flights) {
  console.log(f.flight_number, f.origin, f.destination, f.scheduled_departure);
}

const m = (await getFlightMetrics()).data.data;
// on_time_percentage is over measured_flights, not total_flights.
console.log(`${m.on_time} on time of ${m.measured_flights} measured, `
          + `${m.total_flights} scheduled`);

Without the client, plain fetch works the same way with the header set.

Null is not zero. A null delay_minutes means no time was reported, not that the flight was on time. A null gate_conflicts means no flight carried both a gate and a schedule, so the check could not run. Averaging nulls as zeros understates delay on exactly the flights you know least about.

Endpoint

Enter an API key, or sign in to send requests as yourself.

Sent only in the X-API-Key header of this request. Never stored, and never shown in the code below.

Each request counts toward your plan’s allowance, like any other call.

Request

Response

Send a request to see the response here.

Every record the API returns is a verified observation, labeled with where it came from. A value the source did not report stays null, so you always know what was measured.

data_confidence	Means
`live_feed`	Observed. Came off a provider response.
`database`	Recorded history for this flight.

Each flight list reports in excluded_generated how many records were held back because they could not be verified against a source, so a short result can be told apart from a thin feed rather than leaving you to guess which you have.

To use one source only, filter on data_confidence in your own code. A model trained on live readings should not be reading recorded history alongside them.

A quota you discover only by being refused is not a quota, it is a surprise. Every response carries your position, so a client can slow down in advance.

Response{
  "success": true,
  "data": { },
  "message": "Operation successful",
  "quota": {
    "calls_remaining": 41283,
    "percent_used": 17.4,
    "rate_limit_remaining": 487
  }
}

Mirrored in headers, for clients that do not parse the body: X-Quota-Remaining, X-Quota-Percent-Used, X-RateLimit-Remaining.

GET /api/v1/usage returns the breakdown by endpoint. Call it at startup, not on a timer: the position is already on every response, and polling spends the allowance it reports.

data_confidence and risk_level arrive as typed unions, so a typo fails to compile rather than silently never matching.

Python

pip install kquika-ssystem

Install kquika-ssystem, import ssystem. Python 3.9 or later.

TypeScript

npm install @kquika-inc/s-system

Browser and Node 18 or later. No fetch polyfill.

Both are MIT licensed. Building in another language? The specification is OpenAPI 3.1 and any generator will read it.

API rate limits vary by subscription plan:

Plan	Rate Limit	Burst Limit
Standard	100 requests/minute	200 requests/minute
Professional	500 requests/minute	1,000 requests/minute
Enterprise	2,000 requests/minute	5,000 requests/minute

Subscription Plans

Starter

For small regional airlines

Basic Flight Dashboard
Basic Passenger Insights
Email Support (48h response)

Request Access

Standard

For regional airlines

Flight Operations Dashboard
Passenger Analytics
Email Support (24h response)

Request Access

Popular

Professional

For mid-size to large airlines

Everything in Standard
Predictive Intelligence Suite
Airport sensor and survey readings
Priority 24/7 Support

Request Access

Enterprise

Full-scale solution for major carriers

Everything in Professional
Dedicated Account Manager
Custom Model Training
On-premise Deployment Option

Contact Sales

Support Resources

Comprehensive guides, tutorials, and API references.

• Getting Started Guide
• API Reference Documentation
• Integration Tutorials
• Best Practices

Technical Support

Expert assistance when you need it.

• Email: contact@kquika.com
• Ticket-based support system
• Priority support for Professional+
• 24/7 support for Enterprise

Training

Learn to maximize your S-System investment.

• Online training modules
• Webinar sessions
• On-site training (Enterprise)
• Certification programs

Need Additional Help?

Our team of airline operations experts is ready to provide personalized assistance with implementation, integration, and optimization.

Contact Our Team

S-System Documentation

System Overview

Platform Capabilities

Key Features

Flight Operations Dashboard

Passenger Analytics

Airport Operations

Predictive Analytics

Business Intelligence

Dynamic Offer Pricing

Disruption Management

Predictive Intelligence Suite

How S-System turns your operation into forecasts

Disruption Prediction Engine

Delay Prediction Engine

Customer Intelligence Engine

Passenger Flow Engine

Demand Forecasting Engine

Revenue Optimization Engine

Commercial Forecasting Engine

Real-Time Data Integration

System Architecture

S-System Architecture Overview

Access

Operations modules

Predictive intelligence

Integration

Your Data

Compliance and Assurance

Passenger rights, nine jurisdictions

Published prediction accuracy

Audit trail

Your data stays yours

Dynamic Offer Pricing

Priced from the moment

We price the offer, not the passenger

Revenue you can defend

API Documentation

API Overview

Authentication

Response Format

Flight Operations API

Operations API

Passenger Analytics API

Airport Operations API

Calling the API

Try it

Every row tells you where it came from

Know your quota before you hit it

Client libraries

Rate Limits

Subscription Plans

Starter

Standard

Professional

Enterprise

Support Resources

Documentation

Technical Support

Training

Need Additional Help?