DESIGN AND IMPLEMENTATION OF AN AI-BASED DECISION SUPPORT SYSTEM FOR IT SUPPORT UNITS

Abstract

Artificial Intelligence (AI) is a key driver of digital transformation in organizational processes. When integrated into Decision Support Systems (DSS), it enables faster, data-driven, and more consistent decision-making, especially in operational settings such as IT support units. This study presents the design and implementation of an AI-powered DSS that combines a Natural Language Processing (NLP)–based chatbot with structured data analytics to improve problem resolution, reduce human workload, and generate managerial insights. The system was implemented as a modular web application using the ASP.NET Core MVC framework. A chatbot interface interacts with users in natural language, processes queries through OpenAI’s language-model API, and records sessions in a Microsoft SQL Server database. At the end of each support session, users provide feedback, and the system can generate an automatic summary of the solution process. Administrators access these records through a dashboard that presents analytical visualizations, frequently encountered problems, and performance indicators for chatbot-assisted resolutions. A scenario-based simulation with 50 typical IT support cases was conducted to evaluate the system. Results showed that 88% of responses were considered helpful, while the average time-to-first-suggestion was 1.68 ± 0.21 seconds (client side). AI-generated summaries were also evaluated by human reviewers (Cohen’s κ = 0.87) and found to be coherent and contextually accurate. These findings indicate that AI-integrated DSS architectures can enhance user satisfaction, reduce response time, and support organizational learning by transforming tacit problem-solving interactions into analyzable knowledge assets.