|
Software Metrics Research
|
The availability of
faster, smaller and cheaper hardware has made software an important critical
path component in many products.
Many software projects
are delivered late, over budget and under quality
- or simply cancelled.
Through the analysis
of software project data, we can develop a better understanding of the
key factors which influence software costs and quality.
Cost
Estimation
Key Results
-
Company Specific models are
more accurate than general models.
-
Organizations
should collect their own software metrics data.
-
They should analyze
the data to determine their key productivity factors.
-
They should develop company
specific models based on these factors.
-
Companies can benefit from contributing
to multi-company databases as general models may be more accurate than
guessing and intuition.
Related Research Papers
Maxwell, K., and P. Forselius,
"Software Maintenance Cost Drivers", in Applied Statistics
for Software Managers,
Prentice Hall PTR, June 2002, Abstract
Maxwell, K., L. Van Wassenhove
and S. Dutta,“Performance Evaluation of General and Company
Specific Models in Software Development Effort Estimation,” Management
Science,
June 1999. Abstract
Briand, L.C., K. El Emam,
K. Maxwell, D. Surmann and I. Wieczorek, “An Assessment and
Comparison of Common Software Cost Estimation Modeling Techniques”,
Proceedings of the International Conference on Software Engineering, Los
Angeles, May 1999. Abstract
Productivity
Benchmarking
Key Results
Summary of Results of Software
Development Productivity by Business Sector in 1999
|
Business Sector
|
Productivity
(Function Points/Hour)
|
| Manufacturing |
0.34
|
| Retail/Wholesale |
0.25
|
| Public Administration |
0.23
|
| Banking |
0.12
|
| Insurance |
0.12
|
-
Highest productivity in Manufacturing
Sector.
-
Lowest productivity in Banking
and Insurance Sector.
-
In each business sector, productivity
was explained by a small number of different variables.
-
Productivity
equations give better estimates than business sector averages.
Organizations can benefit from
contributing to multi-company databases through:
Related Research Papers
Maxwell, K., "Collecting
Data for Comparability: Benchmarking Software Development Productivity",
IEEE Software, September-October 2001. Abstract
Maxwell, K. and P. Forselius,
“Benchmarking Software Development Productivity”,
IEEE Software, January-February 2000. Abstract.
Maxwell, K., L. Van Wassenhove
and S. Dutta, "Software Development Productivity of European
Space, Military and Industrial Applications," IEEE Transactions on Software
Engineering, October 1996. Abstract.
Maxwell, K., “Benchmarking
software development productivity: Statistical analysis by business
sector,” Project Control for 2000 and Beyond, Edited by R. Kusters, A.
Cowderoy,
F. Heemstra and J. Trienekens, Proceedings of ESCOM-ENCRESS 98, Rome, Italy,
May 1998, Shaker Publishing B.V.
Maxwell, K., L. Van Wassenhove
and S. Dutta, “Benchmarking : The Data Contribution Dilemma,”
Proceedings of the 1997 European Software Control and Metrics Conference,
Berlin,
Germany, May 1997
Maxwell, K., L. Van Wassenhove
and S. Dutta, “Analysis of European Software Development
Productivity,” Proceedings of the 4th European Conference on Information
Systems,
Lisbon, Portugal, July 1996.
Contact
Dr. Maxwell for copies of papers not available on the website.
Order
Now
|
Applied Statistics
for Software Managers is the first complete guide to using statistical
techniques to solve specific software development and maintenance problems.
You don't need a mathematical background: Katrina Maxwell
presents an easy-to-follow methodology and detailed case studies that show
you exactly how to assess productivity, time to market, development effort
and maintenance cost drivers.
Book includes real software development
and maintenance data from one bank.
Click
here to learn more!
|
