[Paper Review] Validation Measures in CMMI

Mahmoud Khraiwesh|arXiv (Cornell University)|Mar 19, 2011

Software Engineering Research20 references3 citations

TL;DR

This paper proposes a systematic approach to defining validation measures in the CMMI framework using the Goal-Question-Metric (GQM) paradigm. It identifies specific, practical metrics for each goal and practice in the CMMI Validation Process Area, enhancing the framework's applicability for software quality assessment and process improvement in real-world settings.

ABSTRACT

Validation is one of the software engineering disciplines that help build quality into software. The major objective of software validation process is to determine that the software performs its intended functions correctly and provide information about its quality and reliability. This paper identifies general measures for the specific goals and its specific practices of Validation Process Area (PA) in Capability Maturity Model Integration (CMMI). CMMI is developed by Software Engineering Institute (SEI). CMMI is a framework for improvement and assessment of a software development process. CMMI needs a measurement program that is practical. The method we used to define the measures is to apply the Goal Question Metrics (GQM) paradigm to the specific goals and its specific practices of Validation Process Area in CMMI.

Motivation & Objective

• To address the lack of standardized, practical measurement practices in the CMMI Validation Process Area.
• To improve the assessment and maturity of software development processes by defining measurable validation goals.
• To support organizations in implementing effective validation activities through data-driven metrics.
• To bridge the gap between CMMI's theoretical framework and real-world software engineering practice.

Proposed method

• Application of the Goal-Question-Metric (GQM) paradigm to derive metrics from specific goals and practices in the CMMI Validation Process Area.
• Identification of key validation objectives and mapping them to measurable questions.
• Definition of concrete, actionable metrics for each specific practice in the validation process area.
• Use of a structured, hierarchical approach to ensure alignment between goals, questions, and measurable outcomes.
• Focus on practicality and relevance to real software development environments.
• Validation of the proposed metrics through alignment with CMMI's process improvement principles and industry standards.

Experimental results

Research questions

• RQ1 How can meaningful and practical metrics be defined for the validation process goals in CMMI?
• RQ2 What specific questions and measurable indicators support the assessment of validation effectiveness in software development?
• RQ3 How can the GQM paradigm be effectively applied to the CMMI Validation Process Area to ensure traceability and coherence?
• RQ4 What are the key performance indicators that reflect the quality and reliability of software validation activities?
• RQ5 How do the proposed metrics enhance the maturity and assessability of the validation process in CMMI?

Key findings

• The study successfully derived a comprehensive set of metrics aligned with each specific goal and practice in the CMMI Validation Process Area.
• The GQM-based approach enabled the creation of measurable, traceable, and actionable validation indicators.
• The proposed metrics support both process assessment and continuous improvement in software validation.
• The framework enhances the practicality of CMMI by providing clear, data-driven guidance for validation activities.
• The metrics are designed to be implementable in real-world software organizations, improving measurement reliability and process transparency.
• The outcome is a validated, systematic method for measuring software validation effectiveness within the CMMI framework.