AI-Based Predictive quality model for customer defects

• Machine learning techniques can significantly improve the efficiency and accuracy of the customer complaint management process.
• When integrated with AI and predictive models, Quality 4.0 principles contribute to more data-driven decision-making and process automation.
• The study demonstrates the potential of predictive models in identifying responsibility for complaints, leading to reduced analysis time and improved operational efficiency.

In the context of digital transformation and Industry 4.0, this study explores integrating machine learning (ML) techniques with Quality 4.0 principles to enhance the customer complaint management process. By analyzing real customer complaint data from an automotive company, the study aimed to develop predictive models to anticipate responsibility for complaints, ultimately reducing the time and resources needed for analysis. The study’s methodology included ten distinct phases, from data exploration to model evaluation, ensuring the results were applicable across different fields.

The findings revealed that the ML models, particularly XGBoost, achieved an accuracy of 59.7% in predicting complaint accountability, which was later improved to 64% using AutoML techniques. This improvement highlights the importance of feature selection and the need for continuous refinement of predictive models. The study also emphasized the critical role of data centralization and automation in enhancing process efficiency, making information more accessible, and reducing unnecessary complexity in the analysis.

This work underscores the transformative potential of integrating AI techniques into quality management, aligning with Quality 4.0 principles. The study provides a valuable proof of concept, demonstrating how data-driven approaches can significantly improve complaint-handling processes, reduce costs, and optimize decision-making. Additionally, the methodology developed in this study can be replicated in other areas of the company and similar industries, offering a scalable solution for quality management challenges in the digital age.

Read the full study