Using Quantitative Software Metrics to Support CMMI Deployment and Appraisal

• Main Authors: Chun-Yen Chen, 陳俊彥
• Other Authors: Sun-Jen Huang
• Format: Others
• Language: zh-TW
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/31985512407352214829

碩士 === 國立臺灣科技大學 === 資訊管理系 === 91 === While a software organization adopts CMMI model to initiate its software process improvement, a series of activities such as gap analysis, process development, piloting, deployment, mini-appraisal, formal appraisal need to be taken. Formal Appraisal leaded by SEI authorized Lead appraiser (LA) is performed by the regulations of the Appraisal Requirements for CMMI (ARC) and Standard CMMI Appraisal Method for Process Improvement (SCAMPI). However, the assessment criteria, scale labels and ratings defined in both documents for the evaluation of process area’s goals and practices are all qualitative descriptions. Thus, so far CMMI deployment and appraisal activities are all based on the subjective judgment made by appraisal team members or lead appraisers. This thesis aims at proposing quantitative software metric sets for supporting CMMI deployment and appraisal activities. The software metrics proposed in this thesis are derived with references of ARC and SCAMPI standards, ISO 9126 standard for defining the metrics specification, the Practical Software and System measurement methodology for defining the software metrics, and Function Point Analysis (FPA) model for benchmarking the software metrics. This thesis also proposes a two-stage performance assessment and improvement process. The process first uses general metrics for assessing the general practices (GP) within a process area (PA), then using specified metrics for assessing its specified practices (SP). The merit of the quantitative software metric sets proposed in this thesis is to improve the efficiency of CMMI deployment and appraisal activities by providing more objective and valuable information.