A Descriptive Evaluation of Automated Software Cost-Estimation Models,
Abstract
The purpose of this report is to encourage and facilitate the use of appropriate software cost and schedule models in the Department of Defense. This report provides an introduction to the usefulness of these models and an evaluation of current, fully supported, commercially available, software cost and schedule estimation products. These products can be employed for analyses of several different aspects of software management including: 1) Project planning - to arrive at an early estimate of resource requirements and identify tradeoffs between cost and schedule; 2) Proposal evaluation - to assess the realism of contract proposals and assist in negotiation; 3) Managing development - to track resources, get early warning of deviation from plan, and assess alternatives for corrective action; 4) Managing maintenance - to estimate and track the cost of maintenance and enhancement and identify appropriate timing for a new start; and 5) Improving software productivity - to evaluate methodologies and tools which may increase productivity. An important advantage of these parametric models is that they make visible the many assumptions about the product and the development organization that enter into an estimate. These assumptions are therefore open to discussion, validation, and assessment of alternatives. This visibility provides a basis for assessing risk, evaluating reasonableness of a plan, and taking management action or negotiating a bid. In summary, software cost and schedule models form a basis for sharpening software management skills.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 1986
- Accession Number
- ADA178400
Entities
People
- Elizabeth K. Bailey
- John W. Bailey
- Thomas P. Frazier
Organizations
- Institute for Defense Analyses