Home // PATTERNS 2022, The Fourteenth International Conference on Pervasive Patterns and Applications // View article

Concept of an Inference Procedure for Fault Detection in Production Planning

Authors:
Jan Michael Spoor
Jens Weber
Simon Hagemann
Frederik Simon Bäumer

Keywords: Production Planning, Fault Detection, Knowledge Engineering, Data Mining, Case-Based Reasoning

Abstract:
To date, no implemented solution in manufacturing, i.e., in automotive industry, exists to support production planning with insights from production. A structured feedback loop from operations to planning is required to further improve production planning. This contribution discusses the limitations of an existing concept for an inference procedure from operations to new planning tasks using the findings from previous implementation studies. Using the constraints found in these studies, six principles for inference procedures are derived. Thus, the existing concept is renewed and a structured and specific approach in providing an inference procedure for planning activities of similar manufacturing systems is proposed. This approach is split into the different sub-tasks of data acquisition, fault detection, knowledge representation, and knowledge inference. Each sub-task has its unique state-of-the-art solutions, challenges, and limitations which have to be examined during further implementations. Most notably, the concept requires a definition of a normal model to derive fault events and error patterns, an embedding of the fault events in an ontology to create a knowledge base, and the definition of a metric to measure similarity between the current configuration in operation and new configurations of the production planning.

Pages: 10 to 17

Copyright: Copyright (c) IARIA, 2022

Publication date: April 24, 2022

Published in: conference

ISSN: 2308-3557

ISBN: 978-1-61208-953-9

Location: Barcelona, Spain

Dates: from April 24, 2022 to April 28, 2022