Date of Award

7-12-1994

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Science

First Advisor

Dr. Hossein Saiedian

Abstract

For several years, the software industry has been affected by a problem called the software crisis. Many software developers have worked hard to determine remedies for this crisis. The crisis deals with products being produced over costs, behind schedule, with low quality, and not meeting customer satisfaction. Processes are being used which waste resources and build low quality products. In dealing with the software crisis, software practitioners have used many tools, methodologies, and metrics to help produce better products, save resources, and increase productivity. Metrics are meaningful measures used to determine how well software is being produced, where weaknesses are in products, processes, or resources, and indicate where to make improvements. Today, the most popular methodology is the Object-Oriented (00) methodology. Users of this methodology want to prove that it helps resolve the software crisis. There is a belief that 00 systems are developed earlier, easier to understand and maintain, and contain reusable components. However, without metrics, the effectiveness of the 00 approach cannot be proven. Metrics must be used to show if the process or resources being used is responsible for the success or failure of software development, if management of the process is responsible, if the complexity or size of the product being produced is responsible, or if a lack of communication or misunderstandings of requirements is responsible. There are many, many reasons why metrics are used and can be seen in this Thesis. Many metrics are being used for traditional software development methodologies that deal with processes, products, and resources. However, there is no metric paradigm for the 00 development methodology. But, there is a small set of metrics proposed for 00 programming. We provide a 00 metric paradigm that contains metrics grouped into nine categories. We grouped the metrics into categories to make the selection of a particular· metric easier. These nine metrics fit nicely into the 00 methodology. The metrics can be used to measure processes, products, and resources. We provide a size estimation method that helps determine progress, costs, quality, effort, complexity, and schedule. The metrics can be used to prove the effectiveness of the 00 methodology and indicate areas for further improvement.

Comments

A Thesis Presented to the Department of Computer Science and the Faculty of the Graduate College University of Nebraska In Partial Fulfillment of the Requirements for the Degree Master of Science University of Nebraska at Omaha. Copyright 1994 Michael F. Bridges

Download

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

COinS