A Machine Learning-Driven Six Sigma Framework for Enhancing Quality Improvement and Productivity in Aircraft Manufacturing

Aug 5, 2026

The aviation industry faces sustained pressure to raise productivity and efficiency while meeting strict quality requirements. Production defects can create large financial losses through costly rework, delays, and safety hazards. This study proposes a machine learning–driven Six Sigma framework combined with the QCDSME method to improve productivity and analyze quality control in aircraft manufacturing. The DMAIC implementation is referenced to the Airbus A320, using sigma and DPMO metrics, fishbone root-cause analysis, and text mining to determine defect-prone components and human-related factors. Results show defects decrease and sigma improves.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages16

Reading Time40 min

AccessFree to download

Abstract

Introduction

Proposed Framework and Methodology

Six Sigma (DMAIC) Application

Machine Learning and Text Mining Approach

Expected Outcomes

A machine learning-driven Six Sigma framework for enhancing the quality improvement and productivity in the Aircraft Manufacturing Dwi Adi Purnama1*, Alfiqra2, Winda Nur Cahyo1 1Departement of Industrial Engineering, Universitas Islam Indonesia, Yogyakarta, 55584, Indonesia 2PT Astra Otoparts Tbk, Kelapa Gading, Jakarta, 14250, Indonesia *[Corresponding Author](Corresponding Author: dwiadipurnama@uii.ac.id)[:](Corresponding Author: dwiadipurnama@uii.ac.id)[ ](Corresponding Author: dwiadipurnama@uii.ac.id)[dwiadipurnama@uii.ac.id](Corresponding Author: dwiadipurnama@uii.ac.id) | Article history: | | ABSTRACT | | --- | --- | --- | | Received: 2 December 2024
Revised: 25 June 2025
Accepted: 29 June 2025
Published: 30 June 2025
Keywords:
Aircraft manufacturing Quality improvement Six sigma
Machine learning | | The aviation industry, a pillar of global transportation, is under constant pressure to increase productivity and efficiency while maintaining strict quality requirements. Airctraft defects in production can result in significant financial losses, lead to costly rework, delays, and even safety risks. This study proposes a framework to improve productivity and efficiency in aircraft manufacturing and analyze quality control using machine learning, Six Sigma, and the QCDSME (Quality-Cost-Delivery-SafetyMorale) method. The DMAIC (Define-Measure-AnalyzeImprove-Control) stage is a reference in the implementation steps of the Six Sigma method of the Airbus A320. The sigma value in this study was obtained on average for 40 periods of 4.61 sigma and a DPMO of 1225.69. At the analyze stage, a fishbone diagram is used to find the root cause of the problem. Furthermore, a machine learning analysis was performed using the text mining method to identify the most common product components that frequently have defects in Airbus A320 and identify the main factors causing defects, by the human factor. The enhance stage suggests a rise in overcoming challenges with the QCDSME method. Overall, it was discovered that the number of defects fell while the sigma improved and this method can enhance industry performance. | | DOI:
[https://doi.org/10.31315/opsi.v18i1.13960](https://doi.org/10.31315/opsi.v18i1.13960) | This is an open access article under the CC–BY license.
| | 1. INTRODUCTION The aviation industry, a cornerstone of global transportation, faces relentless pressure to enhance productivity and efficiency while maintaining stringent quality standards. As highlighted by Bhatia et al. [1], the aircraft production need to flexibility, adaptability, and data-driven decision-making has become increasingly apparent. Stringent safety regulations, complex supply chains, and the need for continuous innovation have intensified the pressure on manufacturers to deliver high-quality aircraft on time and within budget. The complex nature of aircraft manufacturing involves numerous processes, components, and suppliers, making it difficult to optimize the overall production flow. Inefficiencies can result in increased costs, longer lead times, and reduced customer satisfaction. The aviation industry faces a critical challenge in ensuring the quality and reliability of aircraft products. Defects and low-quality components can lead to serious consequences, including safety risks, operational disruptions, and increased maintenance costs. According to a study by the International Civil Aviation Organization (ICAO), aircraft defects can result in significant financial losses for airlines, with an estimated global cost of over $100 billion annually. Defects and non-conformances can lead to costly rework, delays, and even safety risks. For instance, a study by the European Aviation Safety Agency (EASA) found that manufacturing errors are the leading cause of aircraft defects, accounting for approximately 40% of all incidents. Conventional quality control methodologies, such as manual inspections and basic statisti

FAQs

What problem does the proposed framework address in aircraft manufacturing?

It targets persistent quality issues that reduce productivity by causing defects, rework, delays, and safety risks.

How is Six Sigma implemented in the framework?

The study uses the DMAIC cycle (Define-Measure-Analyze-Improve-Control) as a reference for implementation steps in the Airbus A320 context.

What role does machine learning play in identifying defects?

Machine learning analysis with text mining is used to identify frequently defective product components and the main human-related factors causing those defects.

Aug 5, 2026

Upload by Seraphina