Estimating the total number of defective units is an important problem in quality control and industrial applications. This study develops approximate confidence intervals for the defective total within the framework of the Seemingly Unrelated Regression (SUR) model, where the total is expressed as a linear combination of reliable estimates which yields sum of the estimated percentage of the various components. Using the multivariate central limit theorem, the asymptotic distribution of the estimator vector is derived, leading to the construction of confidence intervals based on both the standard normal (z) and Student’s t quantiles. A simulation study is conducted under different sample sizes and parameter settings to evaluate the coverage performance of the proposed intervals. The results indicate that confidence intervals based on z-quantiles provide satisfactory coverage for moderate and large sample sizes, whereas those based on t-quantiles are generally more conservative, particularly for small samples and low parameter values. The proposed approach offers a practical method for inference on defective totals in correlated multivariate settings................. Keywords: ................SUR model, binomial model, multivariate central limit theorem, asymptotic inference, confidence intervals.