1. Software Engineering
Staffing Level Estimation
Submitted by,
M. Kavitha,
II – M.Sc (CS&IT),
Nadar Saraswathi College
of Arts & Science, Theni.
2. Staffing level estimation :
* The number of personnel required throughout a
software development project is not constant.
* Planning and analysis are performed by a small group
of people, architectural design by a large, but still group and
details design a large number of people.
* Implementation and system testing require the largest
number of people.
* Maintenance may require numerous personnel but the
number should decrease in a short time.
3. * In 1958, Norden research and development project.
Cycles :
A. Planning
B. Design
C. Prototype
D. Product
E. Modification
4. * A cycle of planning, design, prototype, development
and use with the corresponding personnel.
* Any particular point on the Rayleigh curve represents
the number of full-time equivalent personnel at that instants in
time (NOR58).
* The Rayleigh curve is specified by two parameters, td
the time at which the curve its maximum value.
* Then K the total area under the curve which represents
the total effort required for the project.
5. E = K / t2
d t e –t2/2t2
d
Fig 2 : The Rayleigh curve of effort vs. time
6. In 1976, Putnum reported that the personnel level of
effort required throughout the life cycle.
Fig 3 : Putnum’s interpretation of Rayleigh curve(PUT76)
7. A. Design and coding (15%)
B. Test and validation (20%)
C. Extension (10%)
D. Modification (25%)
E. Maintenance (20%)
F. Management (10%)
* The planning requirements analysis and functional
design in the project curve.
* Boehm the Rayleigh is reasonably accurate
estimator of personnel development cycle.
* The development cycle from architectural design
through implementation and system testing if the portion of
the curve between 0.3td and 1.7td.
8. Where ,
PM - estimated number of programmer months
for product development (excluding planning and analysis)
TDEV - estimated development time.
FSP – Full – time Software personnel.
9. * FSP required at any particular time t, where t is
in the range 0.3td to 1.7td.
* Then td is still the time of peak staff requirements
but it is no longer interpreted as the elapsed development
time.
* The plot of the personnel requirement as a
function of time for a 32-KDSI, 91-PM.
COCOMO Rayleigh
Design System
test
FSP
t, months
10. Activity
Effort Schedule
32 KDSI 128 KDSI 32 KDSI 128 KDSI
Plans and
requirements 6% 6% 12% 13%
Architectural
design 16% 16% 19% 19%
Detailed design 24% 24% Combined value :
Coding and unit
test 38% 36% 55% 51%
System test 22% 25% 26% 30%
Table 1 : Distribution of effort for application programs
12. * Boehm also presents tables that specify the of
effort and schedule in a software development project.
* The total number of programmer months and total
development time can be used estimate the actual number
of programmer months and elapsed time for each activity.
* An estimate of the number of full time software
personnel required in each phase of software development
and divided number of programmer month.