Neural network modeling for a two-stage production process with versatile variables: Predictive analysis for above-average performance

With growing academic interest and pragmatic need, adaptive two-stage production modeling becomes an emergent research topic for decision sciences and production management. Although prior research has addressed sequential production process, the primary focus was limited to efficiency analysis with a narrow scope of applications. Data envelopment analysis (DEA) has been commonly used for earlier studies; however, its lack of learning and deficiency in predictive capability seriously diminish the practical utility of DEA and call for an intelligent information-processing technique for further advancement. This paper uniquely presents an output-focused backpropagation neural network (BPNN) approach with capabilities to capture patterns of high performers, a significant departure from conventional efficiency-driven DEA analysis, as well as a promising analytic paradigm. In so doing, the proposed standalone BPNN can predict above-average performance and supports managerial decision-making in setting progressive performance targets in consecutive stages. The sound empirical application to the two-stage bank production process proves the effectiveness of the proposed analytic paradigm. In brief, the intelligent learning model advances existing two-stage production modeling with a methodological breakthrough and makes significant contributions to the existing literature.