Logit Analysis based Evaluation Model for Reliability of Software Company
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
The existing technical evaluation methods for software development depend on the project plan to be realized, not making use of past performance record of the company. The primary factors for a project’s success are not quantitative ones like that of technical expertise, but rather qualitative ones like achievement ability, human resources, infrastructure, and customer confidence. These important factors that evaluate a software company’s reliability are being excluded from the company’s evaluations, making it necessary for devising an evaluation method that can reflect reliability at the time of the company evaluation. This study proposes a model to evaluate software companies based on Logit analysis that can evaluate reliability based on past project performance record of the company at the time of the evaluation. For the evaluation model, interviews and questionnaires were carried out based on software reliability engineering and previous study on project success factors. The GQM-based reliability estimation metric was designed from the results. In order to develop the company evaluation model that uses the defined reliability estimation metric, Logit analysis uses four assessment factors based on the statistical analysis on six assessment factors. The company assessment model is proposed from the results of the Logit analysis. The fitness of the proposed model is verified by a statistical analysis.
Abstract
The existing technical evaluation methods for software development depend on the project plan to be realized, not making use of past performance record of the company. The primary factors for a project’s success are not quantitative ones like that of technical expertise, but rather qualitative ones like achievement ability, human resources, infrastructure, and customer confidence. These important factors that evaluate a software company’s reliability are being excluded from the company’s evaluations, making it necessary for devising an evaluation method that can reflect reliability at the time of the company evaluation. This study proposes a model to evaluate software companies based on Logit analysis that can evaluate reliability based on past project performance record of the company at the time of the evaluation. For the evaluation model, interviews and questionnaires were carried out based on software reliability engineering and previous study on project success factors. The GQM-based reliability estimation metric was designed from the results. In order to develop the company evaluation model that uses the defined reliability estimation metric, Logit analysis uses four assessment factors based on the statistical analysis on six assessment factors. The company assessment model is proposed from the results of the Logit analysis. The fitness of the proposed model is verified by a statistical analysis.
참고문헌 (10)
- Standish Group International, Inc., "Extreme Chaos", ~ (2001)
- Ministry of Information and Communication, "Technical Assessment Standard of Software", ~ (2006)
- Michell, V., "The IT outsourcing marketplace:Vendors and their selection", Jounal of Inform ation Technology, 12(3), pp.223~227 (1997)
- McCall, J., "Methodology for Software Reliability Prediction", RADC-TR-87-171, ~ (1987)
- Neufelder, A. M., "Ensuring Software Reliability" (1992), Marcel Dekker Incorporated
- Jelinski, Z., "Software Reliability Rese arch, In Statistical Computer Performance Evaluation", pp.465~484 (1972), Academic Press
- Goel, A. L., "Time Dependent Error Detection Rate Model for Software Reliability and Other Performance Measures", IEEE Trans. Relibility, R-28, pp.206~211 (1979)
- Lyu, M. R., "The Handbook of Software Reliability Engineering" (1995), McGraw-Hill
- Differding, C., "GQM Tech nology Package" (1996), University of Kaiserslautern
- Lewis, J. R., "IBM computer usability satisfaction questionnaires: psychometric evaluation and instructions for use", International Journal of Human-Computer Interaction, 7(1), pp.57~78 (1995)