In the manufacturing section, due to limitations of specific resources (e.g., time, people, and equipment), key determinants such as process capacity, human resources supply, and equipment availability may be in uncertain or out-of-control environments, followed by decreasing production performance. Traditionally, earlier studies of related issues of production performance usually used statistical methods for handling these problems. However, these methods become more complex when relationships in the input/output dataset are nonlinear. Furthermore, statistical techniques rely on the restrictive assumption on linear separability, multivariate normality and independence of the predictive variables; unfortunately, many of the common models of production performance violate these assumptions. To remedy these existing shortcomings, the study proposes a hybrid procedure that focuses on the opinions of experts, discretization of decision attributes, and application of well-known artificial intelligent (AI) approaches, such as decision trees (DT), artificial neural networks (ANN), and DT+ANN techniques, for objectively classifying production performance to solve real-world problems that are faced by the automobile parts industry. Two practically collected datasets are employed to verify the proposed hybrid procedure. The experimental results reveal that the proposed hybrid procedure is a good alternative to classify production performance from an intelligent manufacturing perspective objectively. Moreover, the output that is created by the DT C4.5 algorithm is a set of comprehensible and meaningful rules applied readily in knowledge-based performance-evaluating systems for manufacturing managers and HR division managers. The study findings and implications are of value to academicians and practitioners.