A generative AI-supported spiral model for developing an MBA dissertation proposal: a proposal

A generative AI-supported spiral model for developing an MBA dissertation proposal: a proposal

Highlight 5 main ideas of the spiral model in software engineering and then discuss how these 5 ideas are also useful (with some adaptations) for guiding the generative AI-supported MBA dissertation proposal development.

 

The spiral model, introduced by Barry Boehm, offers a risk-driven, iterative framework for complex projects. Its five main ideas provide structured flexibility, which adapts well to generative AI-supported MBA dissertation proposal development by incorporating AI for rapid ideation, validation, and refinement.

Main Ideas

·        Risk-driven process: Development prioritizes identifying and mitigating high risks early through analysis, rather than fixed plans.

·        Iterative cycles (spirals): The process repeats in loops, each building on the prior one to refine objectives, design, and outputs incrementally.

·        Objective setting: Each cycle starts by defining goals, alternatives, constraints, and requirements based on prior feedback.

·        Prototype development: Builds tangible prototypes or models in each cycle to test assumptions and gather real-world insights.

·        Evaluation and planning: Ends each cycle with customer review, progress assessment, and commitment to the next spiral's plan.

Adaptations for MBA Proposals

These ideas translate to dissertation proposals by treating the process as iterative spirals, with generative AI accelerating prototyping and risk checks for topics like leadership or housing studies.

Risk-driven process: Identify risks like topic novelty or data access; use AI to scan literature gaps (e.g., query "AI impact on Singapore REITs risks") and simulate supervisor critiques, adapting from software risks to academic feasibility.

Iterative cycles: Cycle through proposal drafts—start broad, refine via AI-generated outlines—mirroring spirals but with shorter loops (days vs. months) for agile academic pivots.

Objective setting: Define research questions and scope per cycle; AI prompts like "refine MBA proposal on dividend stocks" generate alternatives, incorporating user constraints like Singapore focus.

Prototype development: AI creates draft sections or visuals (e.g., literature maps) as "prototypes" for quick testing, faster than manual writing, reducing early commitment errors.

Evaluation and planning: Review AI outputs with self-assessments or peers, then plan next cycle; AI summarizes feedback, ensuring continual alignment with MBA standards like methodology rigor.

A related diagram

*** use generative AI to produce some dissertation proposal prototypes to facilitate the student-researcher's own critical thinking, reflective and deep learning, and exploratory literature review effort to develop and refine his/ her own dissertation proposal with his/ her own voice and in his/ her own words.

A conversation with a student in my MBA lecture:
Student A: "To summarize, the main message is that generative AI is much better than a human supervisor to support and advice his/ her supervisee's dissertation project work."

The lecturer: "This is not the main message; the main message is effective support on a student's intellectual learning, not on the learning-supported tool per se.; the main message is "intellectual curiosity", "engaged learning" and "human development". It is now up to students to decide what is meaningful for them to do."

** In short, precondition for the discussion of the spiral model is on "intellectual learning" and "the authentic interest to learn" [as a single precondition] on the student's part.

** also study a note on a systematic picture of research project tasks.

        A collection of blog notes on using chatgpt for research purpose.