Dr. Ansar Siddique

PhD Theme/Topic: Software Engineering and DevOps Automation Using Multi-Criteria Decision-Making (MCDM) Techniques and Artificial Intelligence

Supervisor: Dr. Ansar Siddique, Professor
Email: ansar.bulc@bahria.edu.pk
Campus/School/Dept: BULC/CS
RAC Approved Supervisor for Research Areas:

Supervisory Record: 
PhD Produced: 0
PhD Enrolled: 01
MS/MPhil Produced: 10
MS/MPhil Enrolled: 01

Topic Brief Description:
Modern software engineering practices increasingly rely on DevOps, which integrates development and operations to achieve continuous delivery, automation, and rapid deployment. However, software development teams continue to face significant challenges, including toolchain complexity, inconsistent CI/CD performance, sub-optimal pipeline configurations, prioritization trade-offs, and time-consuming manual decision-making. Artificial Intelligence (AI) and Multi-Criteria Decision-Making (MCDM) techniques such as AHP, TOPSIS, and EDAS have emerged as powerful approaches to automate complex engineering decisions. Their integration into DevOps can help optimize pipeline performance, improve software quality, automate release decisions, reduce operational overhead, and enhance system reliability.

This research aims to design a decision-support and automation framework that leverages AI and MCDM to assist DevOps teams in tasks such as automated tool selection, build/test optimization, release readiness assessment, and risk-aware deployment decisions.

The proposed solution will reduce human intervention, improve accuracy of decisions, and increase speed and consistency across the DevOps lifecycle.

Research Objectives/Deliverables:

1. To identify key decision factors affecting DevOps automation, including pipeline performance, code quality metrics, testing coverage, deployment risk, and operational KPIs.
2. To evaluate and integrate MCDM techniques (AHP, TOPSIS, EDAS, etc.) for decision optimization within DevOps processes
3. To design and develop an AI-enabled automation framework for DevOps pipeline optimization
4. To build intelligent models for predicting failures, recommending pipeline configurations, and assessing release readiness
5. To develop a prototype DevOps decision-support system integrating MCDM + AI for tool selection, CI/CD prioritization, test optimization, and deployment decision-making
6. To evaluate the framework in terms of accuracy, decision efficiency, software quality improvement, and time reduction

Research Questions:

1. What critical criteria influence automated decision-making in DevOps pipelines?
2. How can MCDM techniques be effectively used to prioritize DevOps tasks, tools, and configurations?
3. Can AI models accurately predict pipeline failures, release risks, and optimal build/test strategies?
4. How can MCDM and AI be combined to build a comprehensive DevOps automation framework?
5. Does the proposed intelligent framework improve software quality, reduce deployment errors, and increase DevOps efficiency?
6. What level of human intervention is reduced through the proposed solution, and how does this affect overall lifecycle performance?

Candidate’s Eligibility Profile:

1. The applicant must have an MS/MPhil/Equivalent degree in electrical engineering with CGPA > 3.0. Besides, applicants must have a strong background in software engineering, DevOps practices, CI/CD pipelines, automation tools, or cloudnative development
2. Familiarity with AI/ML techniques, data analytics, or intelligent decision-making systems
3. Experience with software development technologies such as Git, Docker, Kubernetes, Jenkins, GitLab CI/CD, or similar DevOps toolchains
4. Understanding of MCDM methods (AHP, TOPSIS, EDAS) will be preferred; otherwise, willingness to learn
5. Ability to work independently and collaboratively in a researchdriven environment
6. Strong communication and technical writing skills are essential
7. Proficiency in spoken and written English is essential. We value independence and responsibility while promoting teamwork and collaboration among colleagues.