The impact of computer-aided software engineering on programmer productivity and system quality.

摘要

The high cost of software production has forced organizations to look for new environments to reduce the time required for the development process. Currently, one highly touted route toward improved programmer productivity and increased system quality is Computer-Aided Software Engineering (CASE): the computer-based automation of system development tasks.;Software developers have automated almost every functional area and level of the organization except their own. CASE is an attempt to automate the production of software; computers are being used to enhance the requirements and analysis, design, coding, implementation and maintenance phases of the software development life cycle.;This research addresses two basic questions: when CASE technologies are used to develop software (1) Is programmer productivity improved? (2) Is system quality increased?;An experimental study with a control group (non-CASE) and a treatment group (CASE) was designed to investigate the effects of CASE usage. The same task, Pascal pretty printer, was developed by both groups: Both the developmental processes and the final projects were studied.;This research makes three contributions to the study of software development. First, to our knowledge, it is the first controlled experiment investigating CASE tools. Second, several metrics were identified that can be used to identify and evaluate programmer productivity.;The third contribution of this research is the quantitative measures to the claims of increased programmer productivity and system quality being made by CASE vendors and others. In this study, programmer productivity increased when CASE technologies were used to design a software system. Also in this study, the quality of the systems improved; more complete systems were developed by the teams that used CASE technologies for system design.;Information technology managers should be encouraged in their quest for increased programmer productivity. A major component of the software crisis is the inability to measure, estimate, and improve programmer productivity. This study indicates that use of CASE tools would improve programmer productivity.