Comparison of estimation capabilities of response surface methodology (RSM) with artificial neural network (ANN) in lipase-catalyzed synthesis of palm-based wax ester
2007

Comparing Methods for Synthesizing Wax Esters

Sample size: 30 publication 10 minutes Evidence: high

Author Information

Author(s): Basri Mahiran, Rahman Raja Noor Zaliha Raja Abd, Ebrahimpour Afshin, Salleh Abu Bakar, Gunawan Erin Ryantin, Rahman Mohd Basyaruddin Abd

Primary Institution: Universiti Putra Malaysia

Hypothesis

Can artificial neural networks (ANN) provide better predictions than response surface methodology (RSM) in the synthesis of wax esters?

Conclusion

The study found that ANN outperformed RSM in predicting the synthesis of wax esters from palm oil and oleyl alcohol.

Supporting Evidence

  • The ANN model achieved a coefficient of determination (R2) of 1 for the training set.
  • The actual experimental yield was 84.6%, which was close to the predicted values from both ANN and RSM.
  • The study highlighted the importance of time, temperature, enzyme amount, and substrate molar ratio in determining wax ester yield.

Takeaway

This study looked at two ways to make wax from palm oil and found that one method, called ANN, was better at guessing how much wax would be made than the other method.

Methodology

The study used a central composite rotatable design (CCRD) for experimental design and compared the performance of RSM and ANN in predicting wax ester yields.

Limitations

The ANN requires a large amount of training data compared to RSM, which can provide useful information from fewer experiments.

Statistical Information

P-Value

0.0001

Statistical Significance

p<0.0001

Digital Object Identifier (DOI)

10.1186/1472-6750-7-53

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication