PharmaMind

Intelligent Pharmacy Workforce Continuity Platform

PharmaMind is a full-stack workforce management system for independent pharmacies. It combines a React/Vite frontend, a Flask REST API backend, SQLite persistence, and seven scikit-learn models that predict staffing shortages, demand, pharmacist retention, and pharmacy closure risk.

---

Architecture

PharmaMind/
├── backend/               Flask API + ML pipeline
│   ├── app.py             All API routes (10 sections)
│   ├── auth.py            JWT auth blueprint
│   ├── models.py          SQLAlchemy ORM models
│   ├── ml_models.py       7 ML models (train + predict)
│   ├── simulators.py      Digital twin / continuity simulator
│   ├── scheduler.py       Greedy shift-assignment optimiser
│   ├── seed.py            DB seeder + synthetic dataset generator
│   └── validate_backend.py  End-to-end smoke test
├── frontend/              React 19 + Vite + Tailwind CSS
│   └── src/
│       ├── App.jsx
│       ├── pages/         AdminDashboard, OwnerDashboard, PharmacistDashboard,
│       │                  AnalyticsReports, Login, PendingVerification
│       └── components/    Navbar, CapsuleLanding
└── README.md

---

Setup

Prerequisites

• Python ≥ 3.14
• Node.js ≥ 18

# 1. Clone the repo
git clone https://github.com/vpadival/PharmaMind.git

# 2. Install dependencies (run from the repository root)
pip install -r requirements.txt

# 3. Configure secrets (REQUIRED)
cp .env.example .env
# Edit .env and set a strong JWT_SECRET_KEY:
#   python -c "import secrets; print(secrets.token_hex(32))"

Backend

# 1. Seed the database and train all ML models
python backend/seed.py

# 2. Start the API server
python backend/app.py
# Runs on http://localhost:5000

Frontend

cd frontend

# 1. Install dependencies
npm install

# 2. Configure API base URL
cp .env.example .env.local
# Edit .env.local if your backend runs on a different host/port

# 3. Start the dev server
npm run dev
# Runs on http://localhost:5173

# Build for production
npm run build

---

Demo credentials

Role	Email	Password
Admin	admin@pharmasphere.ai	admin123
Owner 1	owner1@pharmacy.com	owner123
Owner 2	owner2@pharmacy.com	owner123
Pharmacist 1	pharmacist1@pharma.com	pharma123
Pharmacist 2	pharmacist2@pharma.com	pharma123

---

ML Models

Seven models are trained in seed.py on independently generated synthetic datasets (n = 2,000 samples each) and evaluated with train/test splits:

#	Model	Type	Algorithm	Target
1	Shortage Predictor	Classifier	Random Forest	Will there be a staffing gap?
2	Acceptance Predictor	Classifier	Logistic Regression	Will a pharmacist accept offer?
3	Demand Forecaster	Regressor	Random Forest	How many shifts are needed?
4	Trust Risk	Classifier	Gradient Boosting	Risk of next-shift cancellation
5	Workforce Health	Regressor	Random Forest	Health score (0–100)
6	Retention / Churn	Classifier	Random Forest	Will pharmacist go inactive?
7	Closure Risk	Classifier	Gradient Boosting	Risk of pharmacy closure

Calendar features (month, day-of-week) are cyclically encoded using sin/cos transforms before being passed to any model. Evaluation metrics (accuracy, classification report, confusion matrix, RMSE, R²) are printed during seed.py.

---

Running the smoke test

cd PharmaMind
python validate_backend.py

---

Security notes

• JWT secrets are loaded from environment variables — never hardcoded.
• Admin accounts cannot be self-registered via the public API.
• CORS is restricted to the configured FRONTEND_ORIGIN.
• JWT tokens expire after 12 hours.