Joel Henry

Improving software maintenance at Martin Marietta

Using data collected throughout a major project, the authors apply common statistical methods to quantitatively assess and evaluate improvements in a large contractors software-maintenance process. Results show where improvements are needed; examining the change in statistical results lets you quantitatively evaluate the effectiveness of the improvements. We selected a process-assessment methodology developed by J.E. Henry (1993) that follows Total Quality Management principles and is based on Watts Humphreys Process Maturity Framework. It lets you use a process modeling technique based on control-flow diagrams to define an organizations maintenance process. After collecting process and product data throughout the maintenance process, you analyze it using parametric and nonparametric statistical techniques. The statistical-analysis results and the process model help you assess and guide improvements in the organizations maintenance process. The method uses common statistical tests to quantify relationships among maintenance activities and process and product characteristics. The relationships, in turn, tell you more about the maintenance process and how requirements changes affect the product.<<ETX>>

A quantitative comparison of perfective and corrective software maintenance

This paper presents a quantitative comparison of perfective and corrective software maintenance performed by a large military contractor using a formal program release process. The analysis techniques used in the comparison make use of basic data collected throughout the maintenance process. The data collected allow the impact of performing perfective and corrective maintenance to be quantitatively compared. Both parametric and non-parametric statistical techniques are applied to test relationships between and among process and product data. The results provide valuable information for predicting future process and product characteristics, assessing perfective and corrective maintenance impact, and quantitatively comparing the impact of both types of requirements volatility. The results also support one common rule of thumb, cast some doubt on another, and lead to the formulation of a new one.

Quantitative evaluation of software process improvement

Abstract This article describes statistical analysis techniques and results used to quantitatively evaluate software process improvement. The analysis techniques include linear regression, rank correlation, and χ2 tests that have been successfully used to quantitatively assess the software process of a large military subcontractor. A logical extension of this work is to examine the results of these statistical techniques after process improvement. We perform these investigations by altering original data to reflect varying types and degrees of process improvements and then repeating the statistical analyses. We find that different types of process improvement generate very different statistical results. The techniques and results presented here can be used to evaluate the effectiveness of process improvements and determine where continued process improvement is needed.

Defining and implementing a measurement-based software maintenance process

This paper describes the measurement-based software maintenance process defined and implemented at Lockheed-Martin, Moorestown, NJ. The documented process includes extensive data collection, a tightly controlled but highly accessible database, data analysis techniques supported by software tools, and process assessment and improvement activities. The methods and techniques used are presented in a ‘how to’ fashion so that other organizations can leverage our efforts to define and implement a measurement-based process of their own. Our approach is an evolutionary one, rather than a revolutionary organizational upheaval. We describe the benefits gained from our process, including statistically validated metric results, and the subsequent process improvements implemented. This paper describes solutions to the ‘real-world’ issues faced by an organization which successfully implemented a measurement-based software maintenance process.

Coupling and cohesion in object-oriented design and coding

Unfortunately. dcfiniticms of coupling and cohesion arc ncithel well-acccp~ed IIO~ widely ;q~plicd. These definitions are nccdecl in c-n&r to undc~~sl;~ntl Ihc lrotlct~ffs hclwccn design and code altcrnntivcs ;~ntl subscqucntly adv:~n~~ the qualily of objectoriented design and code. Functionally based definitions based on plohiil vnrinblcs. parameters types, data slruclurcs and mtdularity do not directly :~pply to the clhjcct-oriented p;lrdipm. Coupling nntl cohesion exist in ohjcct-(rrientccl soflwarc hut in ways no1 clearly clcfinccl.