نوع مقاله : پژوهشی
نویسنده
مهندسی صنایع - مدیریت و مهندسی صنایع - دانشگاه صنعتی مالک اشتر - تهران - ایرانZ
چکیده
کلیدواژهها
عنوان مقاله [English]
نویسنده [English]
Given the constraints on available resources and the complexity of defense projects, the development of intelligent decision-support tools for prioritizing military projects has become a strategic necessity. This study aims to design an intelligent system for prioritizing defense projects by integrating Multi-Criteria Decision-Making (MCDM) approaches with machine learning algorithms. First, using either the Analytic Hierarchy Process (AHP) or the Best–Worst Method (BWM) and drawing on expert judgments, the weights of key criteria—including strategic value, risk, cost, and level of indigenous development—were determined. Subsequently, project data were segmented using the K-Means clustering algorithm and classified using a Classification and Regression Tree (CART) model. The proposed model achieved an accuracy of 89% in classifying projects into three priority levels: high, medium, and low. A comparison of the results with the conventional Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) demonstrated that the proposed hybrid model offers greater accuracy, transparency, and adaptability. This framework can serve as a practical decision-support tool for defense managers, enabling them to adopt a data-driven approach to identify projects with the highest strategic returns and optimize the allocation of available resources.
کلیدواژهها [English]
)