How to Select the Right SIH Problem Statement: A 10-Point Framework with AWS
A practical framework for selecting an SIH problem statement by analyzing the problem, users, root cause, existing solutions, gaps, MVP feasibility, differentiation, and real-world impact—with AWS as the technology foundation.
How to Select the Right SIH Problem Statement: A 10-Point Framework with AWS
Choosing the right problem statement is one of the most important decisions a team makes in the Smart India Hackathon (SIH).
A problem statement may sound innovative, but that does not necessarily mean it is the right problem for your team.
Instead of asking:
“Which problem sounds the most impressive?”
Ask:
“Which problem can we understand deeply, solve differently, demonstrate realistically, and create measurable impact with?”
This article presents a 10-point framework for evaluating SIH problem statements.
The 10-Point Framework
Problem → Why → User → RCA → Existing Solutions → Gap → Our Difference → MVP → Feasibility → Impact
- Problem — What exactly are we solving?
Start by converting the official SIH problem statement into simple language.
Ask:
What is actually going wrong?
What is the current difficulty?
What happens because of this problem?
Can we explain the problem in 2–3 sentences?
Do not start with technology.
First understand the problem.
- Why — Why does this problem matter?
A technically complex problem is not automatically an important problem.
Understand:
Why does it need to be solved?
What happens if it remains unsolved?
Who or what is affected?
How frequently does it occur?
A strong problem should have a clear reason for existing.
- User — Who actually faces this problem?
Identify the real users and stakeholders.
They could be:
Government departments
Farmers
Doctors
Engineers
Students
Citizens
Disaster-management teams
Researchers
Then ask:
“Who will actually use our solution?”
If you cannot identify the user, it becomes difficult to design the right product.
- RCA — What is the root cause?
Don't only identify the visible problem.
Perform Root Cause Analysis (RCA).
For example:
Problem: Emergency response is slow.
Possible chain:
Slow response → fragmented information → manual verification → delayed decision-making
Now the team can focus on the actual cause instead of simply building another application.
- Existing Solutions — What already exists?
Before developing anything, research the current ecosystem.
Look at:
Government platforms
Commercial products
Research papers
Open-source projects
APIs
Existing AI/ML solutions
This prevents your team from building something that already exists.
- Gap — What is missing?
This is one of the most important questions.
Ask:
“If solutions already exist, why hasn't the problem been completely solved?”
The gap could be:
Low accuracy
High latency
Expensive infrastructure
Poor accessibility
Limited geographical coverage
Lack of real-time processing
Lack of explainability
Poor integration
No offline capability
Your solution should target this gap.
- Our Difference — What are we doing differently?
Now define your unique contribution.
For example:
Existing solution: Prediction system
Our solution: Real-time prediction + explainability + automated alerts + decision dashboard
Simply saying:
“We use AI.”
is not enough.
Your differentiation should solve a specific weakness in existing approaches.
- MVP — What can we realistically demonstrate?
SIH is a hackathon, so scope is extremely important.
You don't need to build the entire production system.
Define the smallest useful version.
For example:
Data → AWS S3 → Processing → ML Model → API → Dashboard → Alert
If this pipeline works reliably, your team has a strong demonstrable MVP.
- Feasibility — Can our team actually build it?
Evaluate your resources honestly.
Ask:
Do we have the required data?
Can we access the necessary APIs?
Do we have the required technical skills?
Is hardware required?
Can we train the model?
Can we deploy it?
Can six team members divide the work?
Can we complete the MVP within the available time?
A great idea that cannot be demonstrated is weaker than a focused idea that actually works.
- Impact — What changes if we succeed?
Finally, determine the measurable impact.
Instead of saying:
“Our solution will improve disaster management.”
Think in measurable terms:
Reduce response time
Improve prediction accuracy
Reduce manual work
Reduce false alarms
Reduce operational costs
Increase geographical coverage
Improve accessibility
The stronger the measurable impact, the stronger the case for your solution.
Where AWS Fits In
AWS should not be added just because a project needs a cloud platform.
Use AWS where it solves an actual engineering requirement.
A possible architecture could be:
Data Sources → Amazon S3 → Processing → ML Model → Amazon SageMaker → API Gateway → Dashboard
For serverless processing, AWS Lambda can execute code without requiring you to manage servers.
For machine learning workflows, Amazon SageMaker provides tools for building, training, and deploying ML models.
For generative AI applications, Amazon Bedrock can provide access to foundation models and AI capabilities.
For monitoring deployed workloads, Amazon CloudWatch can help collect metrics, logs, and operational information.
The important principle is:
Choose the architecture based on the problem, not the other way around.
A Simple SIH Scorecard
After researching each problem statement, score it from 1–10.
Criterion Question
Problem Do we clearly understand the problem?
Importance Does solving it matter?
User Do we know who needs it?
RCA Do we understand the root cause?
Existing Solutions Have we researched what already exists?
Gap Can we identify what is missing?
Difference Can we build something meaningfully different?
MVP Can we demonstrate it?
Feasibility Do we have the data, skills and technology?
Impact Can we measure the improvement?
A problem statement that scores strongly across all ten areas is usually a better candidate than one that simply sounds futuristic.
Final Decision Framework
Before selecting an SIH problem statement, your team should be able to answer:
Problem → What is wrong?
Why → Why does it matter?
User → Who needs the solution?
RCA → Why does the problem happen?
Existing Solutions → What already exists?
Gap → What is still missing?
Our Difference → What will we do differently?
MVP → What can we actually demonstrate?
Feasibility → Can our team build it?
Impact → What measurable improvement will it create?
Only after answering these questions should you decide the technology stack.
Don't start with “What AI can we build?”
Start with “What problem is worth solving, what is missing today, and what can our team realistically prove?”
That mindset can turn SIH problem selection from a guessing game into an engineering decision.
Enjoyed reading this content? Let the author know!
Your likes, comments, shares, and saves help creators reach more builders.
Loading recommendations
Loading article