The SFE experimental data was analysed using the analysis of
variance (ANOVA) to assess the ‘‘goodness of fit’’, which was listed
in Table 2. As shown in this table, the probability (p) value of the
quadratic equation was less than 0.0001 which implied that the
equation was an excellent fit and was very suitable for the present
experimental data. A statistically significant confidence level at
99.999% was obtained. A ‘‘lack of a fit’’ measures the failure of
the equation to represent data in the experimental domain at
points which are not included in the regression. If the p-values of
the ‘‘lack of fit’’ was smaller than 0.05, the model did not fit the
data well [16]. The probability value of ‘‘lack of fit’’ was 0.0507,
suggesting that the model was acceptable [16]. The coefficient of
determination (R2) obtained from the calculated quadratic equation
should also be taken into consideration when validating the
model. The R2 value at 0.9537 was considered very good, indicating
that this equation adequately represented the real relationship
among the parameters chosen. Furthermore, each experiment
was repeated and the statistical analysis was performed so that
the experimental results can be used to valid the model with confidence
which is consistent with the report from the literature [17].