MOTIVATION

Product defects are not only frustrating to the customer, but also to the manufacturing company. Defective products are not just a considerable cost factor, they can lead to a great image damage as well. Accordingly, defect prevention is of essential economic importance for companies.

Approximately 75 % of all defects are caused during product design or development, but are only discovered and corrected in later phases of production or quality assurance. As a matter of fact, the later a product defect is detected, the higher the costs for rectifying it. Companies must therefore aim to avoid product defects as early as possible in order to minimize subsequent costs. Thus, the findings of quality assurance must be incorporated into the process as early as possible, ideally during product development.

Indeed the data collected in the context of quality assurance has so far remained largely unused, except for the optimization of production processes to some extent. Particularly relevant is the early prevention of defects in industries that are characterized by a creative development process in combination with a large number of product variations and raw materials, such as the textile or furniture industry.

AIM

The aim of the research project Preventive QA is to develop a system for preventive quality assurance in industries with creative product design. The goal is to design and implement quality assurance in a preventive manner by incorporating the findings of quality assurance to product development, so that potential product defects can be identified at an early stage.

Preventive QA uses Machine Learning methods to perform an automated analysis of data from quality controls and product returns to determine the relations between product characteristics and quality. The findings are fed back into the product development process, where an assistance system is used to support the creative designers.

The assistance system will provide recommendations to the product designers in order to optimize potential error costs. It will also provide information on alternatives and economic consequences of a design decision. This enables the designer to achieve an optimal balance between design and quality during the development process and to avoid product defects before they occur.

PROJECT CONSORTIUM

In addition to the AWS Institute, the interdisciplinary project consortium consists of the fashion company Casamoda, the software developers Intex and CDS. In addition, the shoe manufacturer HAIX, as an associated partner, will incorporate experiences into the project and evaluate project results for the footwear industry.

The AWSi will bring its expertise in data analysis and assistance systems to the project. The main task of the research project is the development and implementation of algorithms for the automated analysis of structured and unstructured quality data, along with the development of the assistance system that provides recommendations for the product development process.

In addition, the basic system requirements are determined in cooperation with the project partners and an overall concept is developed. Furthermore, AWSi participates in the piloting of the software system as well as the subsequent evaluation.

CONTACT US!

Your contact person

Dr. Christian Linn

Mail: christian.linn@aws-institut.de

Phone: +49 681 93511-302

FUNDING NOTE

The Project Preventive QA is funded by the Federal Ministry of Education and Research.