How design statistics concepts can improve experimentation in medicinal chemistry

Mager, Peter P.

doi:10.1002/(sici)1098-1128(199709)17:5<453::aid-med2>3.0.co;2-t

Cited by 7 publications

(1 citation statement)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The application of DOE models is known to benefit a wide range of industries, such as Aviation (Yondo et al, 2018;Davis et al, 1996), Service (Antony et al, 2020;Setamanit, 2018), Chemical Engineering (Yoo, 2020;Durakovic & Torlak, 2017;Andersson et al, 1996), Engineering (Schlueter & Geyer, 2018), Environment (Okatia et al, 2016;Yang & Tsai, 1998), Food Technology (Yu et al, 2018;Chen, 1996), Manufacturing (Wesling & Emamjomeh, 1994), Materials (Paulo et al, 2017;Bucher & Loos, 1994), Medical (Yip et al, 2020;Mager, 1997), andMicrobiology (Puente-Massaguer et al, 2020). Within the field of engineering and science, DOE's applications are frequently visible in the materials sector (30 %), mechanical engineering (18%), chemical engineering (17 %), and the industrial sector (13 %).…”

Section: What Is Doe?mentioning

confidence: 99%

Critical Evaluation into the practical utility of the Design of Experiments

Sharma

2021

Engineering Management in Production and Services

View full text Add to dashboard Cite

The research aims to emphasise the relevance of the Design of Experiments (DOE) technique as a reliable method for ensuring efficient use of statistical methods in routine industrial processes. A case study approach with a deductive strategy was used to assess the effectiveness of different DOE methods to achieve the desired objectives. Screening, mid-resolution and high-resolution DOE methods helped identify, characterise, and optimise an experimental variable against the desired output response. A general framework for effective DOE is provided as part of DOE planning, including defining DOE objectives, selection criteria, noise reduction, and application across industries. Overall, various DOE models proved successful in identifying a complicated relationship between experimental variables and output response. However, when ideal DOE models may not be feasible, reducing test run by choosing lower resolution DOE or fewer replicates can still provide important insights into the experimental variables’ impact on output responses.

show abstract