An Experimental Investigation of the Decision Processes of Software Managers.
NAVAL POSTGRADUATE SCHOOL MONTEREY CA
Pagination or Media Count:
In recent years, the Department of Defense has been plagued by cost overruns and schedule slippages in major software development projects. As a result, the performance of managers on software development projects has come under increasing scrutiny. Despite this scrutiny, little attention has been paid to the decision processes of software project managers. The goal of this project was to conduct micro-empirical research into managers decision processes, thereby gaining a greater understanding of the ways in which managers make decisions. The project used a simulation of a real NASA software project and verbal protocol analysis techniques to capture the decision processes of seven professional software project managers. Methods of analysis were adapted and developed to trace and explore the captured protocols. Specifically, each subjects decision processes were traced and compared to a nominal model of the decision environment and compared to other subjects. It was found that, despite their experience, the subjects generally had difficulty developing mental models which accurately captured all of the complex relationships in the software management environment. The quality assurance sub-system proved especially difficult. The observed difficulties with the quality assurance decision would prove detrimental to real-world software projects and should be investigated further. Additionally, it was noted that managers tend to place a greater emphasis on meeting schedule goals than cost. Also, more experienced managers tend to rely on specific data, make fewer changes, and are less able or willing to verbalize their thoughts.
- Administration and Management
- Computer Programming and Software