WSPLS — A new approach towards mixture modeling and accelerated product development

García‐Muñoz, Salvador; Polizzi, Mark A.

doi:10.1016/j.chemolab.2012.03.009

Cited by 28 publications

(10 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Latent variable methods are often used to model the relationships between raw material properties and process performance or product quality in pharmaceutical process development [109,110]. For these applications the interpretability of the model parameters is helpful in terms of developing process understanding [105,[111][112][113].…”

Section: Latent Variable Methodsmentioning

confidence: 99%

Mathematical Tools for the Quantitative Definition of a Design Space

Rogers

Ierapetritou

2016

Methods in Pharmacology and Toxicology

View full text Add to dashboard Cite

This chapter focuses on the application of process modeling to determination of design space for pharmaceutical manufacturing processes. The first two sections define design space and related terms from a regulatory perspective and discuss how these apply in practice during pharmaceutical process development. In Sect. 3, a variety of mathematical techniques that can be used to guide design space development are introduced. An emphasis is placed on the use of feasibility analysis and the relationship between process feasibility and design space. Statistical techniques, including latent variable and Bayesian methods, are also discussed in detail. For methods that have been reported in the literature for pharmaceutical manufacturing applications, an overview of the relevant case studies is provided. If methods have not yet been applied to pharmaceutical processes, potential applications for these techniques are indicated. To illustrate the applicability of these concepts to pharmaceutical manufacturing processes, a brief case study is presented in Sect. 4. A discussion of design space verification and issues related to scale-up is provided in Sect. 5. Finally a brief summary of the concepts presented and their potential role in design space development for pharmaceutical processes is given in Sect. 6. This chapter is intended to provide readers with an understanding of mathematical techniques that can be used during process development to assist in the determination of process design space. An overview of the problem formulation and solution approaches for each method are presented. More detailed mathematical developments for each technique are available in the references provided. Table 1 provides a list of symbols that are used throughout this chapter, while Table 2 describes acronyms used within this chapter.

show abstract

Section: Latent Variable Methodsmentioning

confidence: 99%

Mathematical Tools for the Quantitative Definition of a Design Space

Rogers

Ierapetritou

2016

Methods in Pharmacology and Toxicology

View full text Add to dashboard Cite

show abstract

“…In contrast, PLS regression seeks a transformation that explains variance in both the input and response data while retaining directions in the input space which are useful in predicting the responses of interest . Latent variables models are often used in pharmaceutical development to evaluate the relationships between material properties and process performance or product quality . For these applications, the interpretability of the model parameters is helpful in terms of developing process understanding …”

Section: Methodsmentioning

confidence: 99%

Discrete element reduced‐order modeling of dynamic particulate systems

Rogers

Ierapetritou

2014

AIChE Journal

View full text Add to dashboard Cite

One of the key technical challenges associated with modeling particulate processes is the ongoing need to develop efficient and accurate predictive models. Often the models that best represent solids handling processes, like discrete element method (DEM) models, are computationally expensive to evaluate. In this work, a reduced-order modeling (ROM) methodology is proposed that can represent distributed parameter information, like particle velocity profiles, obtained from high-fidelity (DEM) simulations in a more computationally efficient fashion. The proposed methodology uses principal component analysis (PCA) to reduce the dimensionality of the distributed parameter information, and response surface modeling to map the distributed parameter data to process operating parameters. This PCA-based ROM approach has been used to model velocity trajectories in a continuous convective mixer, to demonstrate its applicability for pharmaceutical process modeling.

show abstract

“…In order to introduce the processing conditions within the modeling framework, Muteki et al [29] further proposed an MB-PLS arrangement where the matrix of the processing conditions is used as a regressor together with matrix RX, i.e., a weighted average of the physical/chemical properties of the raw materials. García-Muñoz and Polizzi [30] proposed the use of the weighted average of the PCA scores of matrix X instead of the weighted average of the physical/chemical properties of the raw materials RX. The proposed weighted-scores PLS (WSPLS) model was shown to be more robust than other approaches when dealing with several raw material datasets in which raw materials with different characteristics are collected, and when handling mixture datasets containing missing data.…”

Section: L-shape Pls and Weighted-scores Plsmentioning

confidence: 99%

Advanced Process Decision Making Using Multivariate Latent Variable Methods

Ottavian

Tomba²,

Barolo

2016

Methods in Pharmacology and Toxicology

View full text Add to dashboard Cite

This chapter is intended to show how latent variable modeling techniques can be used to support\ud several pharmaceutical development and manufacturing activities by exploitation of historical databases\ud deriving from experiments, ongoing manufacturing processes or historical products already developed.\ud Basic theoretical concepts about latent variable modeling and latent variable model inversion are first\ud introduced. Then, some applications are reviewed to show how the pharmaceutical industry can benefit\ud from these modeling techniques to support decision-making activities in process development,\ud formulation design, process scale-up, product transfer, process control, and raw materials acceptability\ud assessment

show abstract

WSPLS — A new approach towards mixture modeling and accelerated product development

Cited by 28 publications

References 16 publications

Mathematical Tools for the Quantitative Definition of a Design Space

Mathematical Tools for the Quantitative Definition of a Design Space

Discrete element reduced‐order modeling of dynamic particulate systems

Advanced Process Decision Making Using Multivariate Latent Variable Methods

Contact Info

Product

Resources

About