Scribd
Upload
33 views4 pages

AI Monitoring Stocking System Project Document

The document outlines a project to develop an AI-enabled monitoring and stocking system for resources and consumables, utilizing mobile and web inputs without IoT hardware. It includes a detailed six-week roadmap for implementation, covering planning, backend setup, data cleaning, AI forecasting, dashboard creation, and feedback integration. Additionally, it provides guidelines, team roles, and a starter kit with essential resources for the project.

Uploaded by

PS2
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
33 views4 pages

AI Monitoring Stocking System Project Document

The document outlines a project to develop an AI-enabled monitoring and stocking system for resources and consumables, utilizing mobile and web inputs without IoT hardware. It includes a detailed six-week roadmap for implementation, covering planning, backend setup, data cleaning, AI forecasting, dashboard creation, and feedback integration. Additionally, it provides guidelines, team roles, and a starter kit with essential resources for the project.

Uploaded by

PS2
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

AI-Enabled Monitoring and Stocking

System for Resources and Consumables


Project Guide Document + Starter Kit

Project Summary
You’re building a smart system that:

- Lets staff enter stock/sales/usage/damage data

- Cleans and stores the data

- Uses AI to forecast restocking needs

- Detects weird stock behaviors (errors, anomalies)

- Sends alerts & reminders

- Shows graphs, reports, dashboards

All without using IoT sensors or hardware. Everything is done via mobile or web inputs.

System Architecture (Based on Kavya's Research)


1. Data Input Layer (Web/Mobile)

2. API Gateway (Flask/Django)

3. Backend Services for validation

4. Centralized DB (MySQL/PostgreSQL)

5. ETL for cleaning, formatting

6. Data Warehouse (Optional: CSV or BigQuery)

7. AI Forecasting (Prophet, LSTM)

8. Alerts + Dashboard Output Layer

9. Feedback & Model Improvement


Roadmap (6-Week Detailed Plan)
WEEK 1: Planning + Interface Mockup

- Create mobile/web form for manual inputs

- Responsive UI with validation

WEEK 2: Backend + DB Setup

- Design tables, setup MySQL/PostgreSQL

- Build APIs and connect forms to DB

WEEK 3: Data Cleaning (ETL Layer)

- Remove duplicates, standardize units

- Generate cleaned dataset

WEEK 4: AI Engine

- Train Prophet model to forecast usage

- Detect anomalies using Isolation Forest

WEEK 5: Dashboard + Alerts

- Show visualizations (Plotly/Streamlit)

- Implement alert logic for stock outs

WEEK 6: Feedback Layer

- Add buttons for staff to rate forecasts

- Use feedback to refine model over time


Do's & Don'ts
✅ Keep DB normalized

✅ Use dummy data early (CSV)

✅ Stick to one AI model

❌ Don’t skip integration tests

❌ Avoid over-complication; manual input only

Short Tricks & Tips


- Use Facebook Prophet for easy forecasting

- Streamlit makes fast responsive dashboards

- Use dropdowns in forms to avoid typos

- Add colored alerts for attention in demo

- Store logs in CSV for debugging

Team Roles Suggestion


- PS2: AI Modeling + Team Leader

- Manoj: Backend & DB

- Kavya: UI + Architecture + Flow

- Aadit: ETL + Alerts

Final Submission Checklist


- Project Report (PDF/DOCX)

- Complete codebase + ReadMe

- Sample data (CSV)

- Screenshots + Dashboard visuals

- PowerPoint + Video Demo (optional)

- Hosted app (Heroku/Render/local)


Starter Kit
✅ Sample Tables: Inventory, StockLogs, Deliveries, Alerts

✅ Python Flask Boilerplate for forms & DB connection

✅ ETL Sample Script (CSV → Cleaned Pandas DF)

✅ Prophet Forecasting Code (.ipynb or .py)

✅ Streamlit Dashboard Template

✅ Alert trigger logic (basic threshold example)

✅ Sample dummy data (30-day CSV)

You might also like