Economic Control of Quality of Manufactured Product.

https://www.ece.lsu.edu/mcu/lawss/add_materials/FaultDetectionPart1.pdf

A Review of Process Fault Detection and Diagnosis Part I : Quantitative Model-Based Methods

A review of process fault detection and diagnosis: Part III: Process history based methods

The Mahalanobis distance

This paper provides an overview and analysis of statistical process monitoring methods for fault detection, identification and reconstruction. Several fault detection indices in the literature are analyzed and unified. Fault reconstruction for both sensor and process faults is presented which extends the traditional missing value replacement method. Fault diagnosis methods that have appeared recently are reviewed. The reconstruction-based approach and the contribution-based approach are analyzed and compared with simulation and industrial examples. The complementary nature of the reconstruction- and contribution-based approaches is highlighted. An industrial example of polyester film process monitoring is given to demonstrate the power of the contribution- and reconstruction-based approaches in a hierarchical monitoring framework. Finally we demonstrate that the reconstruction-based framework provides a convenient way for fault analysis, including fault detectability, reconstructability and identifiability conditions, resolving many theoretical issues in process monitoring. Additional topics are summarized at the end of the paper for future investigation. Copyright © 2003 John Wiley & Sons, Ltd.

Statistical process monitoring: basics and beyond

The biopharmaceutical industry has entered a new era. Attention is now being paid to real-time process monitoring, real-time process control, continuous improvement of processes, and quick product technology transfer. Terms like Quality by Design (QbD), design space, control strategy, process analytical technology, process signature are now used to reflect the current state. Multivariate statistical analysis has played an integral part in several industries, enabling process understanding, utilization of real-time analyzers, and real-time product release. It is therefore appropriate to see it as an integral part of the biopharmaceutical industry effort to address issues like design space, control strategy, real-time process signature monitoring, process understanding, and correct technology transfer. This chapter discusses the fundamentals behind multivariate methods for such purposes.

Multivariate Analysis for Process Understanding, Monitoring, Control, and Optimization of Lyophilization Processes

http://www.nt.ntnu.no/users/skoge/prost/proceedings/adchem09/cd/abstract/188.pdf

Quality by Design in the Pharmaceutical Industry: Process Modelling, Monitoring and Control using Latent Variable Methods

It has been claimed [D. H. Melik and H. S. Fogler, J. Colloid Interface Sci.92, 161 (1983)] that by combining the turbidity ratio and the wavelength exponent, a two-parameter size distribution of the particles in a suspension can be estimated from only two turbidity measurements at two widely separated wavelengths, in other words that one is able to estimate three independent parameters (number concentration of particles and the two parameters of the particle size distribution) from only two measurements. It is shown here that this claim is strictly a result of an approximation that was used, and the solution obtained depends only upon the error of this approximation and upon the measurement errors, and has no relationship to the true particle size distribution. Furthermore it is shown that the turbidity ratio and the wavelength exponent methods are essentially the same technique for particle size determination.

Theodora Kourti

Papers

Multivariate Analysis for Process Understanding, Monitoring, Control, and Optimization of Lyophilization Processes

Quality by Design in the Pharmaceutical Industry: Process Modelling, Monitoring and Control using Latent Variable Methods

Minimum number of turbidity measurements required for the determination of particle size distributions

Measuring Particle Size Distribution of Latex Particles in the Submicrometer Range Using Size-Exclusion Chromatography and Turbidity Spectra

Abnormal Situation Detection, Three Way Data and Projection Methods; Robust Modeling for Industrial Applications