Multicriteria target vector optimization of analytical procedures using a genetic algorithm

Weinke, D.; Lucasius, C.B.; Ehrlich, Melanie; Kateman, G.

doi:10.1016/0003-2670(93)80053-n

Cited by 15 publications

(6 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…relative to the specific sample set of p samples. For this study, equation (5) will be used for the primary calibration samples (RMSEC) using X and y with p = m, the secondary updating set (RMSEM) using M and y M with p = s, and a separate set of secondary samples used for the (7) where respective values in the parenthesis are scaled to range between 0 and 1 inclusive. Three other variations of equation 7were evaluated where RMSEM replaces RMSEC and the corresponding combinations with J replacing the L 2 norm.…”

Section: Model Quality Measuresmentioning

confidence: 99%

“…One approach to optimizing a multipenalty based model using multiple measures of model quality is multicriteria (multiresponse) optimization [1][2][3][4][5][6][7]. Broadly speaking, multicriteria optimization involves "Making a systematic and rational decision of the best alternative among several candidates when multiple (and often conflicting) criteria are present."…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fusion strategies for selecting multiple tuning parameters for multivariate calibration and other penalty based processes: A model updating application for pharmaceutical analysis

Tencate

Kalivas

White

2016

Analytica Chimica Acta

View full text Add to dashboard Cite

New multivariate calibration methods and other processes are being developed that require selection of multiple tuning parameter (penalty) values to form the final model. With one or more tuning parameters, using only one measure of model quality to select final tuning parameter values is not sufficient. Optimization of several model quality measures is challenging. Thus, three fusion ranking methods are investigated for simultaneous assessment of multiple measures of model quality for selecting tuning parameter values. One is a supervised learning fusion rule named sum of ranking differences (SRD). The other two are non-supervised learning processes based on the sum and median operations. The effect of the number of models evaluated on the three fusion rules are also evaluated using three procedures. One procedure uses all models from all possible combinations of the tuning parameters. To reduce the number of models evaluated, an iterative process (only applicable to SRD) is applied and thresholding a model quality measure before applying the fusion rules is also used. A near infrared pharmaceutical data set requiring model updating is used to evaluate the three fusion rules. In this case, calibration of the primary conditions is for the active pharmaceutical ingredient (API) of tablets produced in a laboratory. The secondary conditions for calibration updating is for tablets produced in the full batch setting. Two model updating processes requiring selection of two unique tuning parameter values are studied. One is based on Tikhonov regularization (TR) and the other is a variation of partial least squares (PLS). The three fusion methods are shown to provide equivalent and acceptable results allowing automatic selection of the tuning parameter values. Best tuning parameter values are selected when model quality measures used with the fusion rules are for the small secondary sample set used to form the updated models. In this model updating situation, evaluation of all possible models, thresholding, and iterative SRD performed equivalently for the three fusion rules with TR and PLS performed worse. While the application is model updating, the fusion processes are applicable to other situations requiring selection of multiple tuning parameter values.

show abstract

Section: Model Quality Measuresmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Fusion strategies for selecting multiple tuning parameters for multivariate calibration and other penalty based processes: A model updating application for pharmaceutical analysis

Tencate

Kalivas

White

2016

Analytica Chimica Acta

View full text Add to dashboard Cite

show abstract

“…In the period that goes from the second half of the 1980s to the first half of the 1990s, a few relatively simple and naive MOEAs were introduced. Most of them relied on aggregating functions (mostly linear) [104], lexicographic ordering [51], and target-vector approaches [113]. Most of these MOEAs did not modify their selection mechanism or any other component, except for the definition of the fitness function.…”

Section: The Early Daysmentioning

confidence: 99%

The Pyrenees International Workshop and Summer School on Statistics, Probability and Operations Research : SPO 2009 : Jaca, Spain, September 15-18, 2009

Esteban¹

2010

View full text Add to dashboard Cite

This paper provides an overview of the use of metaheuristics for solving multiobjective optimization problems. The metaheuristics discussed include multi-objective evolutionary algorithms (going from the early approches to the most recent research trends in that area), multi-objective particle swarm optimizers, multi-objective artificial immune systems, multi-objective ant colony systems and multi-objective scatter search. In the final part of the paper, we provide a review of sample applications of multi-objective metaheuristics, and a discussion of some of the topics in which more research is required.

show abstract

“…Um exemplo de algoritmo gené-tico em poliotimização em química analítica é a otimização de distâncias entre picos cromatográficos, como função das condições cromatográficas (composição da fase móvel, temperatura do injetor, valor de pH, etc.) 17 . Pode-se encontrar outro exemplo desse tipo de aplicação na determinação de glicose em sangue humano por fotometria.…”

Section: Aplicação Do Algoritmo Genético Em Químicaunclassified

Algoritmo genético em química

Filho

Poppi

1999

Quím. Nova

View full text Add to dashboard Cite

Recebido em 30/4/98; aceito em 2/9/98 GENETIC ALGORITHM IN CHEMISTRY. Genetic algorithm is an optimization technique based on Darwin evolution theory. In last years its application in chemistry is increasing significantly due the special characteristics for optimization of complex systems. The basic principles and some further modifications implemented to improve its performance are presented, as well as a historical development. A numerical example of a function optimization is also shown to demonstrate how the algorithm works in an optimization process. Finally several chemistry applications realized until now is commented to serve as parameter to future applications in this field.Keywords: chemometrics; optimization; genetic algorithm. DIVULGAÇÃO INTRODUÇÃOEm química muitas vezes surgem situações onde se deve considerar uma série de fatores que exercem influência sobre um determinado processo. Devemos então decidir qual desses fatores, ou em quais níveis (quantidades), que irão produzira a melhor resposta baseada em algum parâmetro pré-estabelecido. Neste caso, deseja-se realizar uma otimização numérica do sistema, ou seja, pretende-se encontrar as condições que irão fornecer a melhor resposta desejada baseada nos parâmetros quantificáveis.Por exemplo, em uma determinada síntese, deseja-se determinar quais valores de temperatura, pH e quantidade de catalisador que irão fazer com que numa determinada reação obtenha-se o maior rendimento possível. Outro exemplo seria encontrar quais as melhores condições operacionais de um equipamento para realizar-se uma análise. Como em cromatografia gasosa, onde seria determinada qual a melhor temperatura da coluna e a melhor vazão do gás de arraste que proporcione a separação dos picos.A utilização de métodos para otimização numérica em quí-mica já é explorada há bastante tempo, podendo ser considerada uma das primeiras aplicações da disciplina que hoje se conhece como quimiometria 1 . A quimiometria, na definição atualmente mais aceita, pode ser considerada como a disciplina que utiliza métodos matemáticos e estatísticos, assim como de lógica matemática, a problemas de origem química.Os métodos normalmente utilizados para a otimização de sistemas químicos podem ser divididos em dois grandes grupos. Tem-se os métodos baseados em gradientes e os que não são. Nos métodos baseados em gradiente 2 explora-se o fato de que a partir de processos derivativos pode-se encontrar os mínimos e máximos. Os métodos que não são baseados em derivadas encontram-se assim no outro grupo, e dentre os mais largamente conhecidos e empregados em química destaca-se o método simplex 3 . Este se baseia no deslocamento do simplex, que é uma figura geométrica, sobre a superfície de resposta para encontrar os máximos e mínimos.Mais recentemente tem surgido na literatura e causado grande impacto na comunidade científica o chamados métodos de inteligência artificial 4 . A inteligência artificial faz parte dos processos que simulam via "software" ou "hardware", o talento dos seres racionais para a represen...

show abstract

Multicriteria target vector optimization of analytical procedures using a genetic algorithm

Cited by 15 publications

References 51 publications

Fusion strategies for selecting multiple tuning parameters for multivariate calibration and other penalty based processes: A model updating application for pharmaceutical analysis

Fusion strategies for selecting multiple tuning parameters for multivariate calibration and other penalty based processes: A model updating application for pharmaceutical analysis

The Pyrenees International Workshop and Summer School on Statistics, Probability and Operations Research : SPO 2009 : Jaca, Spain, September 15-18, 2009

Algoritmo genético em química

Contact Info

Product

Resources

About