This paper explores a Monte Carlo simulation-based approach to solving stochastic discrete optimization problems. The method involves generating random samples and approximating the expected value function using sample averages. The sample average optimization problem is solved iteratively until a stopping criterion is met. The authors discuss the convergence rates and stopping rules of this procedure and present a numerical example of the stochastic knapsack problem. They establish that the sample average approximation (SAA) method converges to the true optimal solution with probability one as the sample size increases, and provide bounds on the convergence rates. The paper also discusses algorithm design, including sample size selection, replication, performance bounds, and post-processing techniques. Numerical tests are conducted to evaluate the method's performance, demonstrating its effectiveness in solving stochastic discrete optimization problems.This paper explores a Monte Carlo simulation-based approach to solving stochastic discrete optimization problems. The method involves generating random samples and approximating the expected value function using sample averages. The sample average optimization problem is solved iteratively until a stopping criterion is met. The authors discuss the convergence rates and stopping rules of this procedure and present a numerical example of the stochastic knapsack problem. They establish that the sample average approximation (SAA) method converges to the true optimal solution with probability one as the sample size increases, and provide bounds on the convergence rates. The paper also discusses algorithm design, including sample size selection, replication, performance bounds, and post-processing techniques. Numerical tests are conducted to evaluate the method's performance, demonstrating its effectiveness in solving stochastic discrete optimization problems.