Artificial Intelligence has significantly expanded the analytical capabilities of modern enterprises, enabling unprecedented advances in prediction, pattern recognition, automation, and information processing. Despite these developments, organizational performance continues to depend less on the ability to generate accurate predictions than on the ability to transform heterogeneous information into effective operational decisions. This distinction exposes a persistent gap between advances in artificial intelligence technologies and the engineering of enterprise decision-making processes. Existing research has extensively explored machine learning, deep learning, reinforcement learning, generative artificial intelligence, and autonomous agents; however, these technologies are commonly investigated as independent computational capabilities rather than as components of an integrated decision ecosystem. As organizations increasingly operate in environments characterized by uncertainty, dynamic constraints, regulatory complexity, and rapidly changing operational conditions, the need for a broader engineering perspective becomes increasingly evident. 

This paper introduces Decision Intelligence Engineering (DIE) as an interdisciplinary engineering framework for designing, integrating, governing, and continuously improving autonomous and human–AI collaborative enterprise decision systems. Rather than positioning artificial intelligence as the final objective, the proposed framework considers predictive models, organizational knowledge, reasoning mechanisms, optimization techniques, operational constraints, execution processes, and continuous learning as interconnected components of a unified decision architecture. The paper argues that enterprise intelligence should be evaluated according to the quality, transparency, adaptability, and organizational impact of the decisions produced rather than the isolated performance of analytical models. 

Beyond its conceptual contribution, the study is informed by professional applications conducted in multiple operational environments, including military personnel performance assessment, industrial artificial intelligence for advanced manufacturing, and AI-supported commercial decision platforms. These applications consistently demonstrated that organizational challenges were rarely caused by insufficient data or inadequate predictive performance. Instead, they originated from the absence of systematic engineering mechanisms capable of transforming distributed analytical intelligence into coordinated operational decisions. The recurring architectural similarities observed across these diverse environments provide practical evidence supporting the proposed framework and illustrate its domain-independent applicability. Building upon these observations, the paper proposes a multilayer Decision Intelligence Architecture that integrates data acquisition, knowledge management, predictive analytics, contextual reasoning, decision formation, operational execution, governance, and continuous feedback within a closed-loop adaptive ecosystem. The framework further emphasizes that enterprise autonomy should not eliminate human expertise but rather establish structured collaboration in which artificial intelligence contributes computational scalability while human decision-makers provide contextual judgment, ethical reasoning, strategic interpretation, and organizational accountability. 

The study contributes to the emerging field of enterprise artificial intelligence by shifting attention from model-centric optimization toward decision-centric engineering. It establishes a conceptual foundation for future research on autonomous enterprise systems, multi-agent decision architectures, AI governance, and human–AI collaborative intelligence, while providing a systematic perspective for organizations seeking to engineer trustworthy, explainable, and continuously improving decision ecosystems.