Abstract—ANFIS employ multi-input single-output (MISO)
system. This paper proposes role of input selection using
neuro-fuzzy approach for the prediction of physical properties
of degradable composites namely melt flow index and density.
Experiments on the physical properties of degradable
composites are currently carried out in the laboratories of
Malaysian Palm Oil Board. ANFIS simulation provides good
alternative method for minimize the time consumed, the cost of
production and the cost of labour on the production of
degradable composites. The results reveal that, all the input
selection give high prediction accuracy by smallest Root Mean
Square Error (RMSE) values and high correlation coefficient
(
R), coefficient of determination (2
R) and adjusted R square
(2
R). These findings give additional evidence that ANFIS has
high ability on input selection in the prediction of degradable
composites.
Index Terms—ANFIS, degradable composites, input selection,
physical properties.
Syamsiah Abu Bakar is with the University Kuala Lumpur Malaysia
France Institute, Malaysia (e-mail: syamsiah@unikl.edu.my).
Rosma Mohd Dom and Ajab Bai Akbarally are with the Universiti
Technology Mara, Malaysia (e-mail: rosma@tmsk.uitm.edu.my,
ajab@tmsk.uitm.edu.my).
Wan Hasamudin Wan Hassan is with the Biomass Technology Centre,
Engineering & Processing Division, MPOB, Malaysia (e-mail:
wanhaswh@mpob.gov.my).
[PDF]
Cite: Syamsiah Abu Bakar, Rosma Mohd Dom, Ajab Bai Akbarally, and Wan Hasamudin Wan Hassan, "Role of Input Selection Prediction of Physical Properties of Degradable Composites Using ANFIS," International Journal of Materials, Mechanics and Manufacturing vol. 3, no. 3, pp. 213-217, 2015.
![[Click]](http://www.ijmmm.org/img/ijmmm_v7n2_inspec%28IET%29_indexscreenshot.jpg){kind=link}