Return to search

Quantitative Assessment Of Software Development Project Management Issues Using Process Simulation With System Dynamics Elements

The complexity of software development projects makes estimation and management very difficult. There is a need for improved cost estimation methods and new models of lifecycle processes other than the common waterfall process. This work has developed a new simulation model of the spiral development lifecycle as well as an approach for using simulation for cost and schedule estimation. The goal is to provide a tool that can analyze the effects of a spiral development process as well as a tool that illustrates the difficulties management faces in forecasting budgets at the beginning of a project which may encourage more realistic approaches to budgetary planning. A new discrete event process model of the incremental spiral development lifecycle approach was developed in order to analyze the effects this development approach has on the estimation process as well as cost and schedule for a project. The input data for the key variables of size, productivity, and defect injection rates in the model was based on analysis of Software Engineering Laboratory data and provided for analysis of the effects of uncertainty in early project estimates. The benefits of combining a separate system dynamics model with a discrete event process models was demonstrated as was the effects of turnover on the cost and schedule for a project. This work includes a major case study of a cancelled NASA software development project that experienced cost and schedule problems throughout its history. Analysis was performed using stochastic simulation with derived probability distributions for key software development factors. A system dynamics model of human resource issues was also combined with the process model to more thoroughly analyze the effects of turnover on a project. This research has demonstrated the benefits of using a simulation model when estimating to allow for more realistic budget and schedule determination including an interval estimate to help focus on the uncertainty of early estimates.

Identiferoai:union.ndltd.org:ucf.edu/oai:stars.library.ucf.edu:etd-1994
Date01 January 2006
CreatorsMizell, Carolyn
PublisherSTARS
Source SetsUniversity of Central Florida
LanguageEnglish
Detected LanguageEnglish
Typetext
Formatapplication/pdf
SourceElectronic Theses and Dissertations

Page generated in 0.0251 seconds