Fault diagnosis and comparing risk for the steel coil manufacturing process using statistical models for binary data

Abstract

Use of advanced statistical models can help industries to design more economical and rational investment plans. Fault detection and diagnosis is an important problem in continuous hot dip galvanizing. Increasingly stringent quality requirements in the automotive industry also require ongoing efforts in process control to make processes more robust. Robust methods for estimating the quality of galvanized steel coils are a tool for the comprehensive monitoring of the performance of the manufacturing process. This study applies different statistical regression models: generalized linear models, generalized additive models and classification trees to estimate the quality of galvanized steel coils on the basis of short time histories. The data, consisting of 48 galvanized steel coils, was divided into sets of conforming and nonconforming coils. Five variables were selected for monitoring the process: steel strip velocity and four bath temperatures.The present paper reports a comparative evaluation of statistical models for binary data using Receiver Operating Characteristic (ROC) curves. A ROC curve is a graph or a technique for visualizing, organizing and selecting classifiers based on their performance. The purpose of this paper is to use them in research to obtain the best model to predict defective steel coil probability. In relation to the work of other authors who only propose goodness of fit statistics, we should highlight one distinctive feature of the methodology presented here, which is the possibility of comparing the different models with ROC graphs which are based on model classification performance. Finally, the results are validated by bootstrap procedures.

Uncertainty analysis of PKL SBLOCA G7.1 test simulation using TRACE with Wilks and GAM surrogate methods

“…The link function I is considered to be the same as a linear function of the predictors, η(x). (Debon and Garcia-Diaz , 2012) GLMs can present problems of bias and over-dispersion, therefore an alternative to these models are GAMs, introduced by (Hastie and Tibshirani , 1986), which are an extension of GLMs in the sense that they adjust nonparametric functions to study the relationship between predictive variables and the response. Non-parametric relationships between response and predictor variables are expressed in terms of smooth functions.…”

Predictive inference for system reliability after common-cause component failures

“…While main developments and applications of ROC methods have mostly been reported in medical statistics, they have been increasingly applied in recent years, including to reliability problems. For example, they have been used for the comparison of different statistical models for failures in water supply networks [23] and steel coil manufacturing [24]. Only basic nonparametric methods for ROC analysis are used in this paper, with necessary adaptation to deal with imprecise probabilities as inputs.…”

Fault diagnosis method based on Petri nets considering service feature of information source devices

“…Petri net and Fuzzy Petri net (FPN) have attracted much attention in fault diagnosis [1][2][3]. They have rigorous mathematical definitions, intuitive reasoning processes that are consistent with the occurrence and development of faults, and have good tolerance for data uncertainty [4].…”

“…Therefore, the diagnostic result of FPN will be related to the reliabilities of relays and breakers. (2) At present, the correct operating rate of a device is an average value of statistical data. While the average value covers the variation of the certainty factor of information from a relay or breaker, it cannot represent actual reliability of a relay or breaker at a time, that makes the confidence degree of fault diagnosis results reduce.…”