In a competitive market, companies need to continuously optimize their production processes, improve product quality and reduce costs in order to maintain or increase market share. Step-by-step decision-making helps enterprises make timely and effective adjustments in the face of competitive pressure. This paper focuses on the decision-making problem in the production process. Firstly, the sampling and testing method is used to determine whether the defective rate of a batch of parts exceeds the nominal value of 10%, then the defective rate of spare parts obeying the binomial distribution is approximated as obeying the normal distribution, and a one-sided hypothesis testing model is established. Finally, the number of samples required in both cases and the critical number of defective parts in this sample are calculated. The actual number of defective products sampled is compared with the critical value to determine whether or not to accept the batch of spare parts. At the same time, in order to make decisions at each stage of the production process. We use 0-1 planning to break down each decision and build a step-by-step decision optimization model to calculate the unit cost of each decision by making decisions from parts to products. From spare parts to semi-finished products, we build a multi-objective planning model to keep the cost of semi-finished products as well as the defective rate as low as possible. Finally Bayesian predictive modeling of sampling and testing method is used to calculate the point estimate of the defective rate.