Elevate Flight Operations with RealTime Analytics

Solution Overview

MongoDB's Real-Time Flight Management solution enables airlines to proactively manage and mitigate costly flight delays. To put the impact into perspective, in 2019 alone, flight delays cost European airlines an average of €4,320 per hour per flight. By leveraging intelligent data handling and immediate response, the solution provides real-time monitoring, predictive insights, and optimized resource allocation. Built on MongoDB's flexible document model and time series collections, and utilizing an event-driven architecture with FastAPI for simulation, GCP PubSub for messaging, Vertex AI for advanced analytics, and Google Cloud Functions for serverless event-driven processing, the solution offers a comprehensive approach to minimizing operational disruptions.

Figure 1. Components of an Event-Driven Architecture

The solution delivers:

• Reduced delay propagation: Real-time monitoring and immediate response capabilities help prevent delays from cascading through flight schedules.

• Optimized resource allocation: Intelligent data analysis enables efficient reallocation of resources during disruptions.

• Enhanced customer experience: Better disruption management leads to improved passenger satisfaction and loyalty.

• Scalable operations: Flexible architecture supports growing data volumes and operational demands.

• Real-time decision making: Immediate access to flight data enables quick and informed operational decisions.

Reference Architectures

MongoDB's Real-Time Flight Management solution implements an event-driven architecture that consists of several key components:

This architecture enables three distinct data flows:

• Operational data flow (blue lines): Handles real-time flight operations

• Analytical data flow (green lines): Manages business intelligence

• Internal connections (black lines): Maintains system integration

Figure 2. Event-Driven Architecture

The solution specifically addresses delay management by providing real-time monitoring, predictive insights, and resource optimization capabilities through this integrated event-driven architecture.

Building the Solution

Prerequisites

• Node.js (v14 or later)

• MongoDB (local or cloud)

• Next.js (v12 or later)

• Google Cloud SDK

import base64
import json
from bson import ObjectId
import os
from pymongo import MongoClient
import functions_framework
# Triggered from a message on a Cloud Pub/Sub topic.
@functions_framework.cloud_event
def postPath(cloud_event):
    # Access URI in secrets
    MONGO_URI = os.environ.get('MONGO_URI', "MONGO_URI is not set.")
    MONGO_DATABASE = os.environ.get('MONGO_DATABASE', "MONGO_DATABASE is not set.")
    MONGO_COLLECTION = os.environ.get('MONGO_COLLECTION', "MONGO_COLLECTION is not set.")
    if not all([MONGO_URI, MONGO_DATABASE, MONGO_COLLECTION]):
        print("Error: One or more environment variables are not set.")
        return
    client = MongoClient(MONGO_URI)
    db = client[MONGO_DATABASE]
    collection = db[MONGO_COLLECTION]
    try:
        # Parse the data in the message
        byte_message = base64.b64decode(cloud_event.data["message"]["data"])
        json_str = byte_message.decode('utf-8')
        data_dict = json.loads(json_str)
        print(data_dict)
        # Work with the dictionary fields
        flight_id = data_dict.get("flight_id")
        initial_path = data_dict.get("initial_path_airps")
        new_path = data_dict.get("new_path_airps")
        disruption_coords = data_dict.get("disruption_coords")
        # Check if all required fields are present
        if not flight_id:
            print("Error: flight_id is missing.")
            return
        if initial_path is None or new_path is None:
            print("Error: One or more path fields are missing.")
            return
        # Find the document in the MongoDB collection and update the path field
        result = collection.update_one(
            {"_id": ObjectId(flight_id)},
            {"$set": {
                "initial_path": initial_path,
                "new_path": new_path,
                "disruption_coords": {
                    "lat": disruption_coords[0],
                    "long": disruption_coords[1]
                }
            }}
        )
        if result.matched_count > 0:
            print("Document successfully updated.")
        else:
            print("No document found with the specified flight_number.")
    except Exception as e:
        print(f"Error updating document: {e}")
    finally:
        client.close()

In the end your app should look like this:

click to enlarge

Figure 3. Flight management dashboard with filters

click to enlarge

Figure 4. Flight route and cost monitoring

This solution provides a scalable, event-driven architecture that enables airlines to manage flight operations efficiently while minimizing the impact of delays and disruptions.

For a detailed, step-by-step guide on implementing this solution, including code samples and specific configuration instructions, visit our GitHub repository.

Key Learnings

• MongoDB's flexible schema and time series collections provide a powerful foundation for flight operations—enabling airlines to handle both pre-flight planning data and real-time telemetry efficiently while adapting to evolving needs through flexible document models.

• Event-driven architecture powered by MongoDB's integration with Google Cloud services (Pub/Sub, Cloud Functions) enables immediate response to flight disruptions by processing real-time data streams, helping airlines minimize delays that can cost an average of €4,320 per hour.

• MongoDB's aggregation pipelines transform operational flight data into actionable insights, allowing airlines to optimize routes and resources while Google Cloud Vertex AI provides predictive analytics for cost and delay management.

• The solution demonstrates MongoDB's versatility in handling diverse data types and workloads—from real-time telemetry collection to analytical processing—while maintaining performance at scale through its distributed architecture and change streams capabilities.

Technologies and Products Used

Authors

• Dr. Humza Akhtar, MongoDB

• Rami Pinto, MongoDB

• Sebastian Rojas Arbulu, MongoDB