Cocomo II Constructive Cost Model Boehm Sybren Deelstra

Cocomo II Constructive Cost Model [Boehm] Sybren Deelstra

Why predicting effort? • Effective planning and control • Selecting the right components • Highlight problem areas

COnstructive COst MOdel II • Estimation model for ‘ 90 s, ‘ 00 • cost database and tools for model improvement • analytic framework, tools and techniques for evaluating technology improvement

Market Driven • • • End-User Programming (US: 55 M) Application Generators (0. 6 M) Application Composition (0. 7 M) System Integration (0. 7 M) Infrastructure (0. 75 M)

Co. Mo II Strategy • • Openness Key structure to market sectors Key in- and outputs to information available Tailorable submodels

User Programming Model • No model • Activity-based estimate

Application Composition • Small teams, few weeks • Based on object points: counts of screens, reports and third-generation language components • Complexity factor

Application Generator, System Integration, Infrastructure • Earliest life cycles: Application Composition Model • Exploration phase: Early Design Model • Actual development and maintenance: Post Architectural Model

Modeling • Cost Factors • Metrics for sizing – Object Points – Unadjusted Function Points – Source Lines of Code (SLOC)

Cost Drivers • Use of Tools • Programmer Capabilities • Analyst Capabilities • Multisite Development • Experience (Language, tools, domain) • • • Reuse Documentation Product complexity. . .

Equations Nominal effort in person months: Diseconomy of scale: Effort multipliers:

Effects of reuse Non-linear: – overhead for assessing, selecting and assimilating component – small modifications generate disproportional large costs New code equivalent:

Tool and more info: http: //sunset. usc. edu/research/COCOMOII/