A Neutrosophic Reversibility Tensor Model for Analyzing Quality Dynamics in Academic Discipline Construction at Local Application-Oriented Universities Based on Artificial Intelligence Technology

This paper presents a new model called the Neutrosophic Reversibility Tensor (NRT). It is designed to measure the real quality of academic programs in local application-oriented universities. Many current evaluation systems rely on fixed indicators like graduation rates or course updates. These indicators are often treated as fully reliable. But in reality, some academic programs show progress only on the surface. Inside, they may be facing serious problems like outdated teaching, weak staff, or unclear goals. This creates a false image of improvement. The NRT model solves this problem by combining three special ideas: (1) Evolution and decline in program development; (2) Uncertainty in academic data, and (3) Logical contradictions between what is said and what actually exists. These ideas come from advanced mathematical logic (neutrosophic theory). Together, they form a four-part system that measures growth, decay, contradiction, and vagueness in academic quality. To test the model, we studied a real Artificial Intelligence degree program at a regional university. We used real data and stepby-step equations to calculate the NRT values over five years. The results clearly showed hidden quality issues that were not visible in the university’s official reports. This research offers a new and powerful way to check academic quality using logical, mathematical, and realistic tools, especially when conditions are changing or unclear.