Paper Type
Complete
Abstract
Manufacturing companies face increasing challenges in handling large, heterogeneous datasets resulting from progressive digitalization of dynamic production systems. While knowledge discovery in databases is widely applied to support decision-making by uncovering insights from such datasets, the data preparation phase remains complex, time-consuming, and costly. This paper proposes a heuristic framework using a genetic algorithm to support automation of the data preparation phase within knowledge discovery processes. The algorithm automatically selects and sequences suitable data preparation methods based on an optimization function, reducing manual effort and improving analytical reliability. A case study applying the proposed approach to analyze quality control measurements with a random forest regressor on real-world manufacturing data demonstrates its effectiveness. The results show that automating data preparation enhances efficiency, accuracy, and adaptability in dynamic industrial environments, offering potential for cost reduction and improved decision support.
Paper Number
1420
Recommended Citation
Hochkamp, Florian; Homuth, Michael; and van der Valk, Hendrik, "Enhancing Manufacturing Knowledge Discovery: Heuristic Data Preparation Framework" (2026). AMCIS 2026 Proceedings. 2.
https://aisel.aisnet.org/amcis2026/scuidt/scuidt/2
Enhancing Manufacturing Knowledge Discovery: Heuristic Data Preparation Framework
Manufacturing companies face increasing challenges in handling large, heterogeneous datasets resulting from progressive digitalization of dynamic production systems. While knowledge discovery in databases is widely applied to support decision-making by uncovering insights from such datasets, the data preparation phase remains complex, time-consuming, and costly. This paper proposes a heuristic framework using a genetic algorithm to support automation of the data preparation phase within knowledge discovery processes. The algorithm automatically selects and sequences suitable data preparation methods based on an optimization function, reducing manual effort and improving analytical reliability. A case study applying the proposed approach to analyze quality control measurements with a random forest regressor on real-world manufacturing data demonstrates its effectiveness. The results show that automating data preparation enhances efficiency, accuracy, and adaptability in dynamic industrial environments, offering potential for cost reduction and improved decision support.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
SCUIDT