Control Charts for Batch Processes

Abstract

Applying control charts successfully to batch processes requires the recognition of some distinguishing features of batch processing versus single‐unit processing. Traditional Shewhart‐type charts can be modified for use in batch‐processing practitioners of statistical process control (SPC); however, entirely different control chart methods and practices will often be required. This article reviews the common features of batch processing and the data available to control chart practitioners from typical batch processes and suggests alternatives to standard Shewhart charts for effective process monitoring and control.

Batch process control and monitoring: a Dual STATIS and Parallel Coordinates (DS-PC) approach

“…Other advantages of the proposed method are that the control and monitoring scheme for batch processes is nonparametric unlike other procedures, which widens its applicability to any case, without stablishing the data distribution. Usually, real data doesn't follow an easy-to-use distribution making it difficult to control and monitor any process (Bersimis et al, 2007;Lewis, 2014).…”

“…The batch is the primary processing unit in which raw materials are loaded into and then a series of transformations take place to yield final products. High-quality products are commonly described by quality characteristics (variables), each of which must be controlled within specifications to maintain customer satisfaction and to describe the process performance as the batch progresses (Lewis, 2014;Niang, Fogliatto, & Saporta, 2013).…”

“…The principal contributions in the literature came with multivariate projection methods, which reduce high dimensional data into low-dimensional 'latent variables'. Statistical projection methods, such as principal component analysis (PCA) and partial least squares (PLS) ideally provide, at some degree of confidence, a representation of the correlation structure within data (Lewis, 2014). As for our concern, approaches to batch monitoring are mostly grounded on multiway partial least squares (MPLS) and multiway principal component analysis (MPCA) (Filho & Luna, 2015;Kourti, 2003).…”

Multivariate statistical process control methods for batch production: a review focused on applications

“…In particular cases, several industries rely on batch processing, including, for example, chemicals, pharmaceuticals, food products, fabrics, metals, bakeries, pulp, and paper manufacturers. In this specific situation, high-quality products are commonly described by quality characteristics (variables), each of which must be controlled within specifications to maintain customer satisfaction and to describe the process performance as the batch progresses (Lewis, 2014).…”

“…As for our concern, first novel approaches to batch monitoring were mostly grounded on multiway partial least squares (MPLS) and multiway principal component analysis (MPCA) , 1995a. From that moment on, several strategies and methods have been developed in order to meet statistical assumptions, such as correlationautocorrelation structure, distribution and parameter free, non-linearity, among others, typical in batch production models (Lewis, 2014).…”