Dashboards Data Model Pipeline Assets Team Setup

Mul Biotech Farms · 2026

An almanac of crops, water & risk across India

A resource-intelligence dashboard covering every Indian state — crop calendars, water footprints, irrigation risk, and structured field research, all built on one clean data model.

Project Ledger Mul Biotech Farms · 2026
States mapped0
Mandi price records0
Crop timeline entries0
Relational tables0
Data issues resolved0

The cropping calendar this dataset is built around

Kharif
JUN · JUL · AUG · SEP · OCT
Rabi
OCT · NOV · DEC · JAN · FEB · MAR
Zaid
MAR · APR · MAY
Kharif — monsoon crops (rice, cotton, maize)
Rabi — winter crops (wheat, mustard, gram)
Zaid — summer crops (melon, fodder)
Perennial — tree & plantation
01 — LIVE REPORTS

Three Power BI dashboards

Select a report below and download the .pbix file — open it in Power BI Desktop to explore the full interactive dossier.

india_crop_water_resource_dashboard.pbix
Power BI · .pbix Download

Crop Water & Resource Dashboard

28 states · Irrigation risk · Water footprint · Crop season calendar

↓ Download .pbix
02 — ARCHITECTURE

One dimension, five facts

A star schema. Every fact table joins back to dim_State on the state name.

Dim
dim_State
State · Region · Contributor · SourceSheet
28rows
Fact
fact_Risks
RiskID · State · RiskType · Detail — 3 bullets per risk type
56rows
Fact
fact_CropTimeline
Crop · Season · Sowing/Harvest periods · month numbers
511rows · 0 blank seasons
Fact
fact_WaterEconomics
Crop · Water footprint (L/kg) · Irrigation rank (1–8)
326rows · 0 blanks
Fact
fact_SWOT
Category · Detail — one 8–12 word headline each
269rows
Fact
fact_AdvantagesConstraints
Type · Detail — sharp phrases under 10 words
760rows

All relationships: dim_State[State] → fact tables · one-to-many

03 — ENGINEERING

From raw sheets to a clean model

Seven stages take multi-contributor research and scraped market data into one normalized workbook.

STAGE 01

Scrape

Selenium headless Chrome pulls commodity prices from 30 state portals, clicking through pagination automatically.

cloud_pipeline.py
STAGE 02

Sync

Scraped CSV uploads to Google Drive via a service account. GitHub Actions runs it on a daily schedule.

.github/workflows/scrape.yml
STAGE 03

Extract

All 29 research sheets parsed into six structured tables, cached with pickle to skip repeat work.

scripts/etl/
STAGE 04

Audit

87 issues found: packed cells, LaTeX artifacts, junk placeholders, 65% blank seasons, missing states.

scripts/fix_junk.py
STAGE 05

Season fill

An 80+ crop knowledge base classifies every blank season into Kharif, Rabi, Zaid, or Perennial.

scripts/season_classifier.py
STAGE 06

Water fill

ICAR-based lookup tables fill 49 blank footprints and 70 blank irrigation ranks.

scripts/water_profiles.py
STAGE 07

Normalize text

Risks trimmed to 3 bullets, SWOT to 8–12 word headlines, advantages to under 10 words.

scripts/text_compress.py
OUTPUT

Clean workbook

Six Excel Table objects, zero blanks, zero junk — ready for Power BI.

Custom_PowerBI_Agriculture_Model.xlsx
04 — REPOSITORY

Everything is version-controlled

Dashboards, datasets, scripts, and docs all live in the repo.

📊india_crop_water_resource_dashboard.pbix

Crop water & resource deep dive. Page 1: state dossier. Page 2: crop economics & water sustainability.

1.3 MB · dashboard/↓ PBIX
📊executive_commodity_analysis.pbix

Executive commodity analysis and state-level market intelligence report.

1.1 MB · dashboard/↓ PBIX
📊commodity_price_intelligence.pbix

Commodity price intelligence and mandi market analytics across states.

1.1 MB · dashboard/↓ PBIX
🗃️Custom_PowerBI_Agriculture_Model.xlsx

The normalized star-schema workbook. Six Excel Table objects, zero blanks across all tables.

~120 KB · data/↓ XLSX
🗂️State Assessment Dashboard Agriculture.xlsx

The original 29-sheet research workbook from all four contributors, before cleaning.

523 KB · data/↓ XLSX
🐍cloud_pipeline.py

Selenium scraper plus Google Drive uploader. Pulls prices from 30 portals and syncs to the cloud.

6 KB · scripts/↓ PY
05 — TEAM

Four researchers, one mentor

State research was split across the team; each of the three dashboards was built by a different contributor.

KP
Kartik Patade
State research · ETL & normalization pipeline · crop water & resource dashboard
NK
Nada Khan
State qualitative research
LD
Lavanya Dive
State research · price intelligence dashboard
MN
Malavika Nair
State research · executive commodity dashboard
SR
Soumodip Atanu Roy
Project Mentor · Mul Biotech Farms
06 — DOCUMENTATION

Run it yourself

Clone, install, scrape, and open the dashboards.

1

Clone the repository

git clone https://github.com/soumodip18/IRBAS-Agricultural-Intelligence-Dashboard.git
cd Automated-Agricultural-Intelligence-Pipeline
2

Install dependencies

pip install selenium webdriver-manager pandas openpyxl \
            google-auth google-api-python-client
3

Add your Google Drive credentials

Create a Google Cloud service account, download its JSON key, and store it as a GitHub Actions secret named GCP_CREDENTIALS. The scraper reads it at runtime.

4

Run the scraper

# scrapes all 30 states, uploads the master CSV to Drive
python scripts/cloud_pipeline.py
5

Open a dashboard

Download a .pbix from dashboard/, open it in Power BI Desktop, then point its source to data/Custom_PowerBI_Agriculture_Model.xlsx under Transform Data → Data Source Settings.

07 — CREDITS & ATTRIBUTION

Built under Mul Biotech Farms

This project is open-access for researchers, policymakers, and agricultural professionals. Please credit the team and organisation when sharing or referencing this work.

Organisation
Mul Biotech Farms
Strategic Agriculture Intelligence Framework · India, 2026
Project Mentor
Soumodip Atanu Roy
Mul Biotech Farms · SequestraBionix
Research & Analytics Team
Kartik Patade
State research · ETL & normalization pipeline · Crop water & resource dashboard
Nada Khan
State qualitative research
Lavanya Dive
State research · Price intelligence dashboard
Malavika Nair
State research · Executive commodity dashboard
Disclaimer & Usage

This dashboard and the underlying datasets were developed as part of the Strategic Agriculture Intelligence Framework under Mul Biotech Farms. Data is sourced from official government publications, peer-reviewed research, and verified agricultural databases. If you share, cite, or build upon this work, please credit Mul Biotech Farms and the contributing team. For research collaborations or data access, reach out via the GitHub repository.

Before you continue

Attribution & Usage Agreement

This project was developed under Mul Biotech Farms by the contributing research team. By accessing, downloading, or sharing this work, you agree to the following terms:

· Credit Mul Biotech Farms and the team in any publication, post, or presentation
· Do not redistribute the data, dashboards, or scripts as your own work
· Pass explicit credits to the contributing researchers when sharing
⚠ Warning

Unauthorized use, plagiarism, or failure to provide explicit attribution constitutes a violation of the usage terms and will result in intellectual property reporting and action.

Register your access