Professional Documents
Culture Documents
314
1. INTRODUCTION
Software cost estimating has been an important but difficult
task since the beginning of the computer era in the 1940s.
As software applications have grown in size and
importance, the need for accuracy in software cost
estimating has grown, too. Since the early 1950s, software
development practitioners and researchers have been trying
to develop methods to estimate software costs and
schedules [Zia et.al, 2011]. Software cost estimation
models have appeared in the literature over the past three
decades. However, the field of software cost estimation is
still in its infancy.
Although different software effort estimation techniques
have been introduced, which are being effectively used in
traditional software development, however the diversity of
new software development methodologies has resulted in a
situation where existing effort prediction models
applicability appears to be limited. Agile software
development provides one such difficulty. This
methodology is based on entirely different concept of
software development which is neither suitable to be
calculated by FP analysis technique nor classical effort
estimation methods can be applied that are specifically
developed
for
sequential
software
development
methodologies.
Agile software development has been attached much
importance as a new software engineering methodology. It
emphasizes on good communication between the
315
316
practice
in
agile
software
RESEARCH PROBLEM
PROPOSED MODEL
2.
3.
317
There are other methods, but these are the two most
common ones. In both cases, the estimates are not produced
in terms of units of time; rather they are merely expressions
of Relative Effort which is a good comparative yardstick.
While both of these methods are effective and widely used,
they do not take into account the underlying elements that
affect effort and uncertainty. We have thus developed a
different model that we find to be very effective. This
model is also consistent with the way we develop rankings
of Stories, Defects and Risk.
3.1 Determining the Effort
There is a multitude of factors that affect our ability to
accurately estimate effort. Accurate estimation requires a
multidimensional view to produce accurate and effective
estimates. The challenge, however, is which dimensions do
we measure? If we were to classify the possibilities using a
SWOT according to Internal vs. External influences, we
can eliminate many of the candidates by simply focusing
our attention on the things over which we have influence
and conversely paying less attention to those that we cant.
We keep the vectors to two so as to keep the process as
simple as possible so that we actually use the process and
dont try to sidestep it because it is too cumbersome. Using
two vectors also maintains a consistency with the other
areas of the methodology.
3.2 Story Size
Story size is an estimate of the relative scale of the work in
terms of actual development effort. Table 1 shows five
values, assigned to different types of user stories according
to their size. Wording of the Guideline description can be
changed by the Team itself or even the criteria can be
redefined.
Table 1. Story Size Scales
Value Guidelines
An extremely large story
Too large to accurately estimate
Should almost certainly be broken down
5
into a set of smaller Stories
May be a candidate for separation into a
new project
4
A very large Story
3
2
318
3.3 Complexity
This is complexity of either or both the requirements of the
Story and or its technical complexity. Complexity
introduces uncertainty to the estimate more complexity
means more uncertainty. Table 2 shows 5 values, assigned
to user stories according to their nature. Like Story Size
table, these guidelines are not fixed. These can be adjusted
by the team itself; however we have categorized them to
accommodate all characteristics of Agile software
development methodology.
Table 2. User Story Complexity Scale.
Value Guidelines
Extremely complex
Many dependencies on other stories, other
systems or subsystems
Represents a skill set or experience that is
important, but absent in the team
Story is difficult to accurately describe
5
Many unknowns
Requires significant refactoring
Requires extensive research
Requires difficult judgment calls
Effects of the Story have significant
impact external to the story itself
Very complex
Multiple dependencies on other stories,
other systems or subsystems
Represents a skill set or experience that is
important, but not strong in the team
Story is somewhat difficult for product
owner to accurately describe
Multiple unknowns
4
Comparatively large amount of refactoring
required
Requires research
Requires senior level programming skills
to complete
Requires somewhat difficult judgment
calls
Effects of the Story have moderate impact
E = ni=1(ES)i
319
Friction Factor
0.98
0.95
Very
Highly
Volatile
0.91
1
1
0.98
0.99
0.94
0.98
0.89
0.96
0.98
0.91
0.85
ii.
Team
Composition
Process
Environmental
Factors
Team Dynamics
= ()
=1
320
Extra
High
0.91
0.96
0.90
0.98
0.98
0.96
0.95
0.98
= ()
=1
T = V Days
n
i=1(ES)i
(Vi)D
n
i=1(ES)i
(Vi)D
Days
WD
Where WD is Work Days per Month.
Months
560679
183451
0.327194348
34821
0.062105055
8736
0.015581108
Rent
14634
0.026100496
Travelling
38279
0.068272577
Furniture
2356
0.004202048
27541
0.049120798
15239
0.027179545
12781
0.022795575
Equipment
0.97
T=
Depreciation
Utility Bills
Copyright &
Licensing
Software Purchase &
Subscription
Repair & Maintenance
8393
0.01496935
Stationary
5782
0.010312496
Marketing
4782
0.008528944
24790
0.044214247
Net Ratio
1.680576587
Other Expenses
Time =
1 (100 )
T
100
1 + (100 )
T
100
Span of Uncertainty = Time Time
Time =
321
= ()
=1
1 (100 )
T
100
1 + (100 )
T
100
3.9 Example
INPUT
N No of User Stories
Work Days per Month (WD)
Monthly Team Salary (TS)
No of Days in one Sprint (Sprint Time)
Units of Effort Completed by the Team in one
Sprint
Estimator Confidence in estimation (CL)
METRICS
Story Size Metric (Table 1)
Story Complexity Metric (Table 2)
Friction Factor Metric (Table 3)
Variable Factor Metric (Table 4)
EVALUATION
Completion time (T) is calculated as
T=
n
i=1(ES)i
(Vi)D
WD
Months
INPUT
No of User Stories = 53
Team Velocity = 51
Sprint Size = 10 Days
No of Working days per Month = 22
Monthly Team Salary = 500000
Confidence Level in Estimation = 90%
Friction Factors
Team Composition
0.95
Process
Environmental Factors
Team Dynamics
Dynamic Force Factors
Expected Team Changes
ES=Complexity x Size
Vi is the Initial or Raw Velocity, calculated as
Team members
outside the project
Personal Issues
D=FR x DF
Where FR is product of all four friction factors (FF),
described in Table, which is calculated as:
Expected Relocation
= ()
=1
Vendors Defect
0.89
0.98
0.85
0.98
0.97
0.94
responsibilities
RESULTS
EFFORT = 300 SP
INITIL VELOCITY = 5.1
FRICTION FACTOR (FR) = 0.704302
DYNAMIC FORCES = 0.76749
DECELRATION = 0.540545
0.98
0.98
0.96
0.98
0.97
0.98
322
VELOCITY = 2.4
TIME = 5.2 MONTHS
COST = 5241671.27 (PAK Rs)
TIME Probable = 5.2 MONTHS
TIME Optimistic = 5.6 MONTHS
TIME Pessimistic = 6.9 MONTHS
COST Probable = 5241671.27 (PAK Rs)
COST Optimistic = 4717504.14 (PAK Rs)
COST Pessimistic = 5765838.40 (PAK Rs)
4.
EXPERIMENTAL ANALYSIS
P.No
Effort
Vi
Sprint
Size
Work
days
Team
Salary
Act:
Time
Est
time
Actual
Cost
Estimated
cost
Time
MRE
Cost
MRE
156
4.2
0.687
2.7
10
22
230000
63
58
1200000
1023207.14
7.93
14.73
202
3.7
0.701
2.5
10
21
260000
92
81
1600000
1680663.89
11.95
5.04
173
0.878
3.3
10
22
250000
56
52
1000000
992269.51
7.14
0.77
331
4.5
0.886
3.8
10
22
300000
86
87
2100000
2002767.22
1.16
4.63
124
4.9
0.903
4.2
10
22
300000
32
29
750000
676081.32
9.375
9.84
339
4.1
0.903
3.6
10
22
400000
91
95
3200000
2895132.85
4.39
9.52
97
4.2
0.859
3.4
10
22
250000
35
29
600000
540113.84
17.14
9.98
257
3.8
0.833
10
22
250000
93
84
1800000
1614078.94
9.67
10.32
84
3.9
0.646
2.4
10
22
190000
36
35
500000
507264.58
2.77
1.45
10
211
4.6
0.758
3.2
10
22
250000
62
66
1200000
1267179.55
6.45
5.59
11
131
4.6
0.758
3.2
10
22
250000
45
41
800000
786732.223
8.88
1.65
12
112
3.9
0.773
2.9
10
22
200000
37
39
650000
597142.61
5.40
8.13
13
101
3.9
0.773
2.9
10
22
200000
32
35
600000
538494.68
9.375
10.25
14
74
3.9
0.773
2.9
10
22
200000
30
26
400000
394545.65
13.33
1.36
15
62
3.9
0.773
2.9
10
22
200000
21
22
350000
330561.22
4.76
5.55
16
289
0.742
2.8
10
22
250000
112
103
2000000
1971485.44
8.03
1.42
17
113
0.742
2.8
10
22
250000
39
40
800000
770857.32
2.56
3.64
18
141
0.742
2.8
10
22
250000
52
50
1000000
961866.44
3.84
3.81
19
213
0.742
2.8
10
22
250000
80
76
1500000
1453032.29
3.13
20
137
3.7
0.758
2.7
10
22
220000
56
51
800000
854347.55
8.92
6.79
21
91
3.7
0.758
2.7
10
22
220000
35
34
550000
567484.33
2.85
3.17
REFERENCES
AGARWAL, R., KUMAR, M., YOGESH, MALLICK, S.,
BHARADWAJ, R. M. & ANANTWAR, D. (2001) Estimating
Software Projects. ACM SIGSOFT Software Engineering Notes,
26, 60-67.
ANGELIS, L., STAMELOS, I. & MORISIO, M. (2001) Building
a Software Cost Estimation Model Based on Categorical Data.
Proceedings of the 7th International Software Metrics Symposium.
BECK, K., BEEDLE, M., VAN BENNEKUM, A., COCKBURN,
A., CUNNINGHAM, W., FOWLER, M.,
HIGHSMITH, J., HUNT, A., GRENNING, J., MELLOR, S.,
JEFFRIES, R., KERN, J., MARICK, B., MARTIN, R. C.,
SCHWABER, K., SUTHERLAND, J. & THOMAS, D. (2001)
The Agile Manifesto.
323
324
Syafadhli A.A.B., Mohamad D. and Sulaiman N.H., DistanceBased Ranking Fuzzy Numbers, Advances in Computational
Mathematics and its Applications (ACMA), Vol. 1, No. 3, 2012;
pp 146-150
TAUSWORTHE, R. C. (1980) The Work Breakdown Structure in
Software Project Management. The Journal of Systems and
Software, 1, 181-186.
Vajargah B. and Jahanbin A., Approximation theory of matrices
based on its low ranking and stochastic computation, Advances in
Computer Science and its Applications (ACSA) Vol. 2, No. 1,
2012; pp 270-280
Vajargah1 B.F., Moradi M. and Kanafchian M., Monte Carlo
optimization for reducing the condition number of ill conditioned
matrices, Advances in Computational Mathematics and its
Applications (ACMA), Vol. 1, No. 1, March 2012; pp 169-173
Viswanadham K.N.S. and Raju Y.S., Quintic B-spline Collocation
Method for Eighth Order Boundary Value Problems, Advances in
Computational Mathematics and its Applications (ACMA),Vol. 1,
No. 1, March 2012; pp 47-52
Yang X., Zhang Y., A New Successive Approximation to Nonhomogeneous Local Fractional Volterra Equation, Advances in
Information Technology and Management (AITM) Vol. 1, No. 3,
2012; pp 138-141