A model of software change risk assessment measures using risk mitigation process in anti-ageing

• Main Author: Azman, Thamaratul Izzah
• Format: Thesis
• Language: English
• Published: 2020
• Subjects: Risk assessment - Case studies; Software maintenance; Aging - Prevention
• Online Access: http://psasir.upm.edu.my/id/eprint/98238/1/FSKTM%202020%2027%20IR.pdf; http://psasir.upm.edu.my/id/eprint/98238/
• Institution: Universiti Putra Malaysia

Continuous changes during maintenance may cause software structure to deteriorate and causing bugs and errors, which reduces software quality leading to software ageing. Software ageing is inevitable, however, the progress of ageing can be delayed to attain software anti-ageing where software maintains its quality. Change analysis involves assessment of risks that monitor, examine and manage the impact of software changes to deal with software ageing, however, it is still unclear how maintainers perform the assessment of risks. Existing risk mitigation is lack of quantifiable approach, which arises ambiguous issues for change analysis. There are also inadequate tools to support maintainers for change analysis during software maintenance for software anti-ageing. The study aims to determine the risks of software changes that influence software ageing from software engineering perspective and further develop a model of software change risk assessment measures using risk mitigation process in anti-ageing as well as building a prototype based on the model. From a comprehensive theoretical study, six risks of software changes such as human, technical, environment, technology, resources and maintenance procedure and process were discovered to influence software ageing during software maintenance. To examine the relationships between those risks and software ageing, a quantitative survey was conducted using a structured questionnaire among 152 software practitioners in Malaysia. The data was analyzed using Structural Equation Modeling (SEM) analysis consists of measurement model and structural model assessment through SmartPLS software. The result shows that human risk, technical risk, environment risk, technology risk and maintenance procedure and process risk have significant effect on software ageing. The study found resources risk has no significant effect on software ageing. The study also discovered that risk mitigation is a moderator for the relationship between software ageing and software anti-ageing, where the interaction term demonstrates a significant path with the p- value of interaction term is 0.0001, which is smaller than recommended p-value of 0.05. Based on the results, a model of software change risk assessment measures using risk mitigation process in anti-ageing is developed that comprises of five components such as change request, risks of software changes, software ageing, risk mitigation process and software anti-ageing. Then, a prototype named Risk Mitigation for Software Anti-Ageing System is built based on the model. The model was validated and prototype was verified using expert or accreditation approach through interview with experts. The findings from this research contributes to assist maintainers to monitor, evaluate and manage risk of software changes that influence software ageing during change analysis in software maintenance to achieve software anti-ageing.