1cost6

An Experiment Measuring the Effects of Personal Software

Process (PSP) Training

Prechelt and Unger

IEEE TOSE May 00

2cost6

Output reliability

3cost6

Surprising data (no digit)

4cost6

Effort estimation error

5cost6

Estimated/actual productivity

6cost6

Productivity

7cost6

Productivity

8cost6

effort

10cost6

An Empirical Study Using Task Assignment Patterns to Improve the Accuracy of Software Effort

EstimationSmith, Hale, Parrish

IEEE TOSE March 01

11cost6

Goal

This paper reports the results of an empirical studyexploring the impact of four task assignment factors, teamsize, concurrency, intensity, and fragmentation, on softwaredevelopment effort. By adding these factors to the IntermediateCOCOMO I model (hereafter simply referred to asCOCOMO), its' estimates are improved significantly for thestudy project.

12cost6

What is basic formula?

E is an effort estimate, expressed in man-months (MM).EDSI refers to the number of Estimated Delivered SourceInstructions. The parameters a and b are determined by theapplication complexity mode; both of these values increasewith application complexity. The effort adjustment factor,EAF, is equal to 1 for the basic level and equals the productof 15 cost factors for the intermediate and advanced levels.

13cost6

Effort Factors

Based on a review of prior research [3], [6], [8], [12], fourmodule assignment factors, team size, concurrency, intensity,and fragmentation, were identified as potentially significantdeterminants of software development effort.

14cost6

Model of Process

15cost6

TEAM

16cost6

INTS

17cost6

CONC

18cost6

PFRAG

19cost6

Models

20cost6

Models

21cost6

Experiment

22cost6

Model Comparison

23cost6

Impact of work asgmnt

24cost6

For Tues 9/24

A Reference Model for Requirements and Specifications

Carl Gunter, Elsa Gunter, Michael Jackson and Pamela Zave

IEEE Software, May/June2000 pp37-43