The nominal-the-best S/N ratio shown in Table 2 is used to determine the optimal level of process parameters namely, barrel  temperature  (A),  mould  temperature  (B),     holding

pressure (C), and injection speed (D) for achieving a target value for the demoulding force in a consistent manner. Hence, the demoulding force achieved at this optimum condition is expected to remain consistently high enough for ejection of moulded parts to achieve a good replication process. The larger-the-better criterion is used to identify the maximum demoulding force and achieve high S/N ratio. The software Minitab 16 was used for analyzing and plotting the results. The results are plotted in Figures 3 and 4 to display the effects of the process parameters upon the demoulding force. The optimum levels for process parameters for nominal-the-best are A2B2C2D2 and for larger-the-better are A2B2C2D1 as indicated in the plots. Both of the set design of experiment is analysed to define the level process parameters from its main factor in micro injection moulding and the confidence level.

Fig. 3. Main effects plot for S/N ratio at nominal-the –best

Fig. 4. Main effects plot for S/N ratio at larger-the –better

The main effect graphs also show the significance of process parameters. The steeper the line the more significant and the graphs show the parameters C and B i.e. holding pressure (Ph) and mould temperature (Tm) are both significant while A and D i.e. barrel temperature (Tb) and injection speed (Vi) are less significant. For nominal-the best the barrel temperature (Tb) is least significant while for larger-the-better it is the injection speed (Vi).

Generally, both graphs show that the parameter B i.e. mould temperature (Tm) must be set at high level in order to fill into the micro cavities more efficiently due to viscosity of the melt polymer. From the nominal-the-best analysis, the mould temperature (Tm) played a main role for contribution in achieving optimum level for demoulding force. Thus, it   can

improve the replication process by keeping the heat to prevent fast cooling in micro cavities. In other words, the cooled layer of melt fills eventually can block the micro cavities. However, the larger-the-better showed that holding pressure (Ph) is most significant due to the fact that the melt fills can be solidified completely before the moulded parts could be ejected. Holding pressure time starts from the injection of melt fills and prevents the melt fills from overflow. The duration of the holding pressure gives sufficient time for melt fills to be frozen in micro cavities of the mould before the melt is cooled completely.

In Taguchi analysis the calculations made by focusing on the values of S/N ratio to determine the level setting of process parameters has shown some interesting and positive results. In general, the larger value for S/N ratio is better for the output results since this measure will ensure that the external factors that can influence the replication process is kept minimal so that the replication capabilities are increased The nominal-the-best criterion will enable the process to achieve and maintain a recommended amount of demoulding force necessary for an optimum production of replication process. Meanwhile, the larger-the-better is most appropriate for high surface quality and mass production.

The interaction between the process parameters are plotted in Figures 5 and 6. The interactions between the parameters indicate if adverse effects are caused by increasing or decreasing one parameter upon the measured response. Parallel lines in the plot shows no interactions while intersecting lines show the presence of interaction.

Fig. 5. Interaction plot for S/N ratio at nominal-the –best

Fig. 6. Interaction plot for S/N ratio at larger-the –better

For nominal-the-best, Fig. 3 shows some interaction between A and C and A and D. The results are optimum when the parameters are at level 2. For larger-the-better, Fig. 4 shows some interaction between B and C and A and D. The results are optimum when the parameters are at level 2.