SIMULATION OF SOFTWARE COST ESTIMATION MODEL USING COCOMO

Accurate software cost estimation is a critical success factor in project management, yet it remains a significant challenge due to the inherent uncertainties in software development. Over-estimates or under-estimates can lead to project failure, resource misallocation, and financial losses. The Constructive Cost Model (COCOMO), developed by Barry Boehm, is one of the most renowned algorithmic models for software cost estimation. This project focuses on the design and implementation of a simulation model for software cost estimation using the COCOMO methodology. The primary objective is to develop a functional software tool that automates the application of the COCOMO model, enabling project managers to input project parameters such as lines of code and cost drivers to obtain reliable estimates for effort (in person-months), schedule (in months), and overall cost. The system was developed using a structured software engineering approach. The front-end was built with HTML, CSS, and JavaScript for an interactive user interface, while the back-end logic implementing the COCOMO I (Basic) and COCOMO II (Post-Architecture) models was developed in PHP. The system allows users to select project modes (Organic, Semi-Detached, Embedded), adjust effort multipliers, and simulate different project scenarios. Testing and validation of the system were conducted using historical project data. The results demonstrate that the simulated model produces estimates that closely align with expectations from the theoretical COCOMO model. The system provides a practical, user-friendly platform for students and project managers to understand and apply COCOMO, thereby bridging the gap between theoretical models and their practical application in software project planning. This work establishes a foundation for future enhancements, including integration with machine learning for parameter calibration and expansion to include other estimation models like Function Point Analysis.