Baggage Weight Prediction Using Machine Learning Segmentation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current baggage management systems in the aviation industry are inefficient and inaccurate in predicting baggage weight, leading to sub-optimal aircraft space management and negative passenger experiences due to congestion and underutilization of checked luggage compartment space.

Innovation Solution

A baggage weight prediction system that uses a processor, data collector, and data analyzer to create a baggage prediction model based on passenger attributes and baggage patterns, incorporating AI and machine learning techniques for real-time data analysis and visualization, enabling accurate segregation of carry-on and checked baggage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If flat average weight per bag is calculated based on standard industry allowance, then baggage management decisions can be made, but prediction accuracy deteriorates and ancillary revenue optimization is reduced

Engineering Contradiction:

Improvebaggage weight prediction accuracyVSAvoidprediction system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system changes the parameters used for baggage weight prediction from simple flat averages to multiple dynamic parameters including passenger demographics, travel purpose, season, weather, and historical data. This transforms the prediction approach to achieve higher accuracy while managing complexity through structured data collection and machine learning models.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent replaces traditional mechanical/manual baggage weighing and estimation methods with an automated machine learning-based prediction system. This substitution uses algorithms that analyze multiple parameters to predict baggage weight, eliminating the need for physical weighing in all cases and providing more accurate predictions than manual estimation.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Productivity

If checked-in luggage compartment space is sub-optimally utilized due to lack of prior information, then baggage management operations can proceed, but aircraft space management efficiency deteriorates

Engineering Contradiction:

Improveaircraft space management efficiencyVSAvoidprior baggage weight and volume information

Core Design Contradiction:

ProductivityVSLoss of information

Solution Approach 1:

The system performs preliminary action by predicting baggage weight and volume before the flight occurs. This advance prediction allows airlines to optimize aircraft loading plans, ensure proper weight distribution, and maximize space utilization in both checked-in luggage compartments and carry-on areas, thereby improving overall aircraft space management efficiency.

Inventive Principle:

Principle #10Preliminary action

3Ease of operation

If carry-on baggage space congestion occurs, then passenger experience deteriorates, but the system lacks capability to predict and prevent it

Engineering Contradiction:

Improvepassenger experienceVSAvoidcarry-on baggage weight prediction

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The system incorporates feedback mechanisms by continuously analyzing actual baggage data from flights and using it to refine prediction models. This feedback loop improves the accuracy of carry-on baggage weight predictions over time, enabling better space management and preventing congestion that negatively impacts passenger experience.

Inventive Principle:

Principle #23Feedback

4Adaptability or versatility

If traditional baggage analysis tools are used, then basic baggage weight calculation is achieved, but strategic decision-making capability for effective baggage management is reduced

Engineering Contradiction:

Improvebaggage management decision capabilityVSAvoidinsights into potential baggage weight

Core Design Contradiction:

Adaptability or versatilityVSLoss of information

Solution Approach 1:

The system segments baggage prediction into different categories including checked-in baggage and carry-on baggage, and further segments by passenger demographics, travel purpose, and environmental factors. This segmentation enables targeted prediction models for different scenarios, providing actionable insights for strategic baggage management decisions rather than providing a single aggregate number.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS11820534B2Baggage weight prediction

Publication Date: 2023.11.21 ACCENTURE GLOBAL SOLUTIONS LTD

AI summary

Examples for the baggage weight prediction system are provided. The system may create a baggage prediction model in response to a carrier baggage weight prediction requirement requesting a total baggage weight to be carried in a future transportation operation by a carrier. The baggage prediction model may be used to determine a predicted baggage to be carried by a plurality of future passengers associated with a future transportation operation. The system may determine a predicted weight for the predicted baggage. The system may sort the predicted baggage into checked baggage and a carry-on baggage. The system may determine a carry-on baggage weight associated with the carry-on baggage based on the baggage prediction model. The system may determine a baggage weight ratio associated with the predicted baggage. The system may generate a baggage prediction result for performing a baggage management action.