Artificial intelligence has started improving several mundane tasks and activities in our day-to-day life. Teaching pedagogies need to undergo several transformations to facilitate the delivery of key takeaways of concepts to solve real-world challenges. The current education system facilitates grasping concepts and applying these concepts to real-world application to find solutions. Work-integrated learning particularly requires participants to design problem statements and then decompose the same into subtasks with required solutions. This may require a clear grasp of concepts which may span across multiple disciplines which have several challenges such as knowledge expertise and infrastructure. To resolve the above, a solution has been proposed which uses artificial intelligence tools like ChatGPT to formulate problem statements based on data repositories available in domains. The model works by choosing a set of broad domains for a use case. Based on this, seed functions which serve as initial basis functions to perform training and create an AI model are designed. Later an appropriate AI model is chosen which operates on the seed functions to create a use case. The AI model analyses the limitations in the respective domains and suggests possible improvement strategies. Each of these suggestions would be a partial optimal solution. Finally, all these partial optimal solutions are aggregated to generate the required use case. The model proposed has been tested with 1000 problem statements, and the performance was assessed. The ChatGPT model provided inferences with an improved reliability of 33% and a reduced fact identification rate of 34%, compared to the conventional manual method.