Artificial Neural Networks and Neuro-Fuzzy Models: Applications in Pharmaceutical Product Development

Singh, Inderbir; Kaur, Jaswinder; Kaur, Sukhanpreet; Barik, Bibhuti Ranjan; Pahwa, Rakesh

doi:10.1590/1678-4324-2023210769

Cited by 1 publication

(1 citation statement)

References 88 publications

(93 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, the wall slip effect may be significant until dominating the perceived behaviour of a product. Under this condition, it becomes a challenging task to perform certain operations for industrial purposes, especially in manufacturing, transportation and designing of materials (Barnes, 2004;Raja et al, 2021;Singh et al, 2023;Wu et al, 2022). For example, the operation of the pumping process for slippery materials cannot be optimised because the wall slip cannot be simply neglected in the selection of a suitable pump.…”

Section: Introductionmentioning

confidence: 99%

Actual Shear Rate Prediction Associated with Wall Slip Phenomenon Using Radial Basis Function Network

Chin,

Lai,

Lee

et al. 2023

IEMJ

View full text Add to dashboard Cite

Wall slip can be defined as a phenomenon where the particles in a suspension moving away from the wall boundary, leaving a thin liquid-rich layer adjacent to the boundary. Such a phenomenon may induce a significant impact on rheological measurements, particularly to viscosity, shear stress, and shear rate. Suspension has wide applications such as food processing, personal care products, pharmaceuticals, paints, medicine, and agrochemical. The traditional technique for the actual shear rate prediction is challenging yet difficult and non-favourable from the perspective of time and cost-effectiveness. Therefore, development of a mathematical computational model which has the ability to perform the prediction task with an acceptable level of accuracy is highly in need. Since radial basis function network (RBFN) has the ability to perform input-output mapping in a highly accurate manner, both of the approaches are employed to generate the actual shear rate prediction model. Through the model evaluation using a series of statistical analyses, it was found that RBFN model V is the best model as it shows the highest coefficient of determination (0.9998), lowest mean squared error (0.001058) and root mean squared error (0.001447), most negative Akaike information criterion (-18426.5) and Bayesian information criterion (-18415.5), and the smallest percentage error.

show abstract

Section: Introductionmentioning

confidence: 99%