From past statistics on the production of an electronic component in a manufacturing plant, it is expected that the average defective rate for the component produced by the assembly line is 20 parts per million. Recently, the plant has implemented a new operating procedure for the assembly of this component. The management expects this new assembly procedure to further reduce the defective rate of the component assembly. In order to justify this improvement, a random sample of 20 shifts of the production line has been selected and the defective rate found in each shift recorded. The sample mean defective rate of the component and its standard deviation are 25 parts per million and 10 parts per million respectively. Assuming the defective rate in each shift of the assembly line is normally distributed, (i) construct 95% confidence interval for mean defective rate to assemble the component using new operating procedure. Interpret the result obtained. Is the new working procedure effective in altering the overall defective rate of the component? (ii) how large a sample shifts of production should be taken if the estimated error for mean defective rate is within 2 parts per million at 95% of confidence level ?