Sample sizes based on exact and approximate approaches to achieve a power of |$80\%$| for |$\theta=0.8$| and |$0.75$|⁠, |$\alpha=0.025$| and keeping |$\lambda_R=21$| and |$\lambda_P=7$| under three different allocations. The simulated power (⁠|$ \hat{\phi} $|⁠) and estimated average type-I error (⁠|$ \hat{\alpha} $|⁠) for exact Bayesian approach under non-informative Gamma prior are also reported to show that calculated sample size is adequate to guarantee |$80\%$| power except for minor numerical fluctuation. Note, Frequentist type-I error is always strictly maintained at |$\alpha= 0.025$| by equation 4.1.