Delphi cost estimation model
Download as PPTX, PDF1 like6,777 views
The Delphi technique was developed to gain expert consensus on estimates without group influence. It can be adapted for software cost estimation by having estimators provide anonymous estimates in rounds. A coordinator summarizes estimates between rounds and asks outliers to justify differences, iterating until consensus. A variation allows group discussion with the coordinator but maintains anonymous estimating to focus on variances. Additional information may be needed if consensus is not reached.
Delphi cost estimation model
- 1. Delphi Cost Estimation Shashwat Shriparv [email protected] InfinitySoft
- 2. The Delphi technique was developed at the RAND corporation in 1948 to gain expert consensus without introducing the adverse side effects of group meetings.
- 3. The Delphi technique can be adapted to software cost estimation in the following manner: •A coordinator provides each estimator with the System Definition document and form for recording a cost estimate. •Estimators study the definition and complete their estimates anonymously. They may ask questions of the coordinator, but they do not discuss their estimates with one another.
- 4. • The coordinator prepares and distributes a summary of the estimators’ responses, and includes any unusual rationales noted by the estimators. • Estimators complete another estimate again anonymously using the results from the previous estimate. Estimators whose estimates differ sharply from the group may be asked, to provide justification for their estimates.
- 5. • The process is iterated for as many rounds as required. No group discussion is allowed during the entire process. • The following approach is a variation on the standard Delphi technique that increases communication while preserving anonymity.
- 6. • The coordinator provides each estimator with a system definition and an estimation form. • The estimators study the definition, and the coordinator calls a group meeting so that estimators can discuss estimation issues with the coordinator and one another. • Estimators complete their estimates anonymously.
- 7. • The coordinator prepares a summary of the estimates but does not record any rationales. • The coordinator calls a group meeting to focus on issues where the estimates vary widely. • Estimators complete another estimate again anonymously. The process is iterated as many rounds as necessary.
- 8. • It is possible that several rounds of estimates will not lead to a consensus estimate. • In this case the coordinator must discuss the issues involved with each estimator to determine the reasons for the differences. • The coordinator may have to gather additional information and present it to the estimators in order to resolve the differences.