Renny Sari Dewi
The early COSMIC theory was applied to estimate software development costs through functional size measurement, based on four types of data movements: Entry, Exit, Read, and Write. There are three artificial intelligence (AI) and internet of things (IoT) projects to test the proposed method. The total COSMIC Function Points (CFP) were 128, 222, and 245. A high coefficient of determination (R2) of 89.07% indicates a strong positive correlation between the functional size and the estimated development cost. Final cost estimations using payrate-based effort modeling resulted in IDR 481,080,000 for Project A, IDR 831,960,000 for Project B, and IDR 918,840,000 for Project C. These findings confirm the reliability of early COSMIC in producing accurate, early-stage cost estimates for software development, particularly in projects contributing to educational quality and technology advancement. Future research should incorporate larger project datasets, compare early COSMIC with other established estimation models, and include additional productivity and risk factors to enhance the robustness and generalizability of cost prediction. © 2025 IEEE.
Department of Digital Business, Universitas Negeri Surabaya, Surabaya, Indonesia
