Information Fusion for Assistance Systems in Production Assessment

TL;DR

This paper introduces a comprehensive framework for information fusion in assistance systems within production assessment, enhancing prediction robustness and uncertainty quantification through evidence theory, validated on industrial and benchmark data.

Contribution

It presents a general evidence theory-based fusion framework for multiple sources and a specific method for combining machine data and expert models, addressing data drift.

Findings

Improved prediction robustness and uncertainty estimation.

Effective fusion of machine data and expert models.

Successful validation on industrial and Tennessee Eastman benchmark data.

Abstract

We propose a novel methodology to define assistance systems that rely on information fusion to combine different sources of information while providing an assessment. The main contribution of this paper is providing a general framework for the fusion of n number of information sources using the evidence theory. The fusion provides a more robust prediction and an associated uncertainty that can be used to assess the prediction likeliness. Moreover, we provide a methodology for the information fusion of two primary sources: an ensemble classifier based on machine data and an expert-centered model. We demonstrate the information fusion approach using data from an industrial setup, which rounds up the application part of this research. Furthermore, we address the problem of data drift by proposing a methodology to update the data-based models using an evidence theory approach. We validate the approach using the Benchmark Tennessee Eastman while doing an ablation study of the model update parameters.