Project/Area Number |
12680347
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Single-year Grants |
Section | 一般 |
Research Field |
計算機科学
|
Research Institution | Osaka University |
Principal Investigator |
KIKUNO Tohru Graduate School of Information Science and Technology, Dept. of Information Systems Engineering, Professor, 大学院・情報科学研究科, 教授 (50093745)
|
Co-Investigator(Kenkyū-buntansha) |
MIZUNO Osamu Graduate School of Information Science and Technology, Dept. of Information Systems Engineering, Assistant Professor, 大学院・情報科学研究科, 助手 (60314407)
|
Project Period (FY) |
2000 – 2002
|
Project Status |
Completed (Fiscal Year 2002)
|
Budget Amount *help |
¥3,200,000 (Direct Cost: ¥3,200,000)
Fiscal Year 2002: ¥900,000 (Direct Cost: ¥900,000)
Fiscal Year 2001: ¥1,100,000 (Direct Cost: ¥1,100,000)
Fiscal Year 2000: ¥1,200,000 (Direct Cost: ¥1,200,000)
|
Keywords | Risk management / Software project / Muitivariate regression model / Project management / Factor analysis / Cluster analysis / WWW / ロジスティック回帰モデル |
Research Abstract |
During the process of software development, project managers often find indications that projects are risky and take appropriate actions to recover them from this dangerous status.If project managers fail to detect such risks, it is possible that such projects may collapse completely. Thus, monitoring technique to avoid such confuse projects becomes necessary requirements for successful software developmet. In this research, we firstly propose a new scheme for the characterization of confused projects based on an evaluation by the project manager. In order to acquire the relevant data to make such an assessment, we designed a questionnaire from five viewpoints. We then analyzed the responses to the questionnaires as provided by project managers by applying a logistic regression analysis. The experimental results using actual project data showed that the proposed characterizing scheme can be the first step toward predicting confused projects at an early phase of the development. Secondly, we investigated the relations between responses of the questionnaire and the cost and the duration of software projects. By the factor analysis, we reconstructed the items of questionnaire into several essential factors that include more important information. Using the reconstructed factors, we can successfully estimate the cost and duration of projects. Thirdly, in order to improve the accuracy of the first approach, we introduce the cluster analysis into the prediction of confused projects. In this research, we proposed a procedure to predict confused projects based on the analogy of the responses of questionnaire. Experimental evaluation showed that the accuracy of prediction is better than that of first approach.
|