TY - CHAP

T1 - A development approach for a standardized quality data model using asset administration shell technology in the context of autonomous quality control loops for manufacturing processes

AU - Bilen, Ali

AU - Stamer, Florian

AU - May, Marvin Carl

AU - Lanza, Gisela

N1 - Conference code: 23

PY - 2023

Y1 - 2023

N2 - In modern markets, increasing quality requirements require high performance quality assurance processes which guarantee the fulfillment of these requirements sustainably. The quality assurance must therefore be able to take targeted countermeasures in the event of deviations. It is becoming increasingly decisive to achieve quality with minimum resource expenditures, increasing the importance of quality control loops. Currently, however, a great deal of manual effort goes into the design and implementation of the mostly knowledge-based control logics due to the heterogeneous data landscape and the resulting data preparation processes. Autonomous quality control loops represent a new development and are intended to provide an efficient and data-based approach to setting up quality control loops. According to the "plug and play"principle, the control system should be operational with a minimum of resources in order to enable precision engineering. Prerequisites for such autonomous systems are homogeneous data structures and models for the holistic representation of quality data, which make individual data preparation processes obsolete. In addition, individual process models must also be replaced by suitable data-based, learning modeling methods. In the following approach, the fundament for a holistic quality data model is developed on the basis of various interviews with diverse companies active in the field of metal-cutting and additive manufacturing. The data model is represented using the Asset Administration Standard of the I4.0 platform. In addition, machine learning approaches in the area of machining and additive manufacturing are analyzed for the general modeling of the correlation between process parameters and the quality result, in order to be able to develop a holistic concept for autonomous quality control on this basis in the next step.

AB - In modern markets, increasing quality requirements require high performance quality assurance processes which guarantee the fulfillment of these requirements sustainably. The quality assurance must therefore be able to take targeted countermeasures in the event of deviations. It is becoming increasingly decisive to achieve quality with minimum resource expenditures, increasing the importance of quality control loops. Currently, however, a great deal of manual effort goes into the design and implementation of the mostly knowledge-based control logics due to the heterogeneous data landscape and the resulting data preparation processes. Autonomous quality control loops represent a new development and are intended to provide an efficient and data-based approach to setting up quality control loops. According to the "plug and play"principle, the control system should be operational with a minimum of resources in order to enable precision engineering. Prerequisites for such autonomous systems are homogeneous data structures and models for the holistic representation of quality data, which make individual data preparation processes obsolete. In addition, individual process models must also be replaced by suitable data-based, learning modeling methods. In the following approach, the fundament for a holistic quality data model is developed on the basis of various interviews with diverse companies active in the field of metal-cutting and additive manufacturing. The data model is represented using the Asset Administration Standard of the I4.0 platform. In addition, machine learning approaches in the area of machining and additive manufacturing are analyzed for the general modeling of the correlation between process parameters and the quality result, in order to be able to develop a holistic concept for autonomous quality control on this basis in the next step.

KW - Correlation Analysis

KW - Data Model

KW - Industrie 4.0

KW - Machine Learning

KW - Quality Control Loops

KW - Quality Data

KW - Engineering

UR - http://www.scopus.com/inward/record.url?scp=85175160567&partnerID=8YFLogxK

M3 - Article in conference proceedings

AN - SCOPUS:85175160567

T3 - European Society for Precision Engineering and Nanotechnology, Conference Proceedings - 23rd International Conference and Exhibition, EUSPEN 2023

SP - 443

EP - 446

BT - European Society for Precision Engineering and Nanotechnology

A2 - Riemer, Oltmann

A2 - Nisbet, Clare

A2 - Phillips, Dishi

PB - euspen

T2 - 23rd International Conference of the European Society for Precision Engineering and Nanotechnology - EUSPEN 2023

Y2 - 12 June 2023 through 16 June 2023

ER -