The increasing demand for transparent, objective, and development-oriented staff performance appraisal in tertiary institutions necessitates the modernization of conventional evaluation systems. This study presents the design and implementation of an AI-based automated staff appraisal system developed for the Federal University of Technology, Owerri (FUTO). The proposed system replaces traditional manual and semi-digital appraisal processes with a web-based platform built using React.js, Node.js, and PostgreSQL. It integrates generative artificial intelligence through prompt-engineered large language models (LLMs) accessed via OpenRouter to generate structured, personalized feedback. A weighted scoring algorithm was implemented to compute performance scores across multiple academic dimensions, including teaching load, research output, professional development, and administrative responsibilities. The system was developed using the Design and Development Research (DDR) methodology, incorporating iterative prototyping, stakeholder consultation, and system validation. Evaluation involved functional testing, performance benchmarking, and user acceptance assessment among academic staff. Results indicate an average AI feedback generation time of 6.3 seconds and high user ratings for usefulness (4.6/5) and ease of use (4.7/5). The system standardizes evaluation criteria, reduces processing delays, and produces structured developmental feedback aligned with institutional performance goals. The architecture demonstrates scalability, modular AI integration, and secure deployment, providing a replicable framework for digital transformation of staff appraisal processes in higher education institutions.
Amadi et al. (Sat,) studied this question.