Cost Estimation Models in Software Engineering
Cost estimation in software engineering involves techniques to predict the financial resources
required for developing and testing software. These models use mathematical algorithms or
parametric equations to estimate costs.
Types of Cost Estimation Techniques
1. Empirical Estimation Technique
Uses empirically derived formulas based on historical data and expert guesses.
Relies on prior project experience and assumptions, often formalized through methods like the
Delphi technique and Expert Judgement technique.
2. Heuristic Technique
A practical approach for solving problems quickly using simple, flexible methods.
Involves shortcuts and rough calculations for quick decision making.
Includes models like the Constructive Cost Model (COCOMO) for estimating costs.
3. Analytical Estimation Technique
Breaks down tasks into basic components for detailed analysis.
Uses standard times or experience based estimates for each component.
Examples include Function Point Analysis (FPA), Putnam Model, Price to Win Estimation, and
machine learning based models like neural networks and regression analysis.
The COCOMO Model
The Constructive Cost Model (COCOMO) is a procedural cost estimation model introduced by
Barry Boehm in 1981. It estimates various project parameters such as size, effort, cost, time, and
quality, based on a study of 63 projects. Key outputs are effort (measured in person months) and
schedule (measured in time units like weeks or months).
