Skip to content

README.md

Latest commit

 

History

History
241 lines (169 loc) · 9.42 KB

README.md

File metadata and controls

241 lines (169 loc) · 9.42 KB

CapEx Factory Readiness Command Center — Reducing Tool Install Delays Through Predictive Readiness Tracking

capex-readiness-ci GitHub Pages Streamlit

Built from 7+ years managing $500M+ CapEx portfolios — A command center approach to de-risk tool installations across NPI programs. Translates fragmented operational data into executive decision-making tools where execution discipline + financial governance + cross-functional coordination intersect.

All data is synthetic/anonymized.

Category Why It Matters Link
Live Dashboard See how I visualize complex program data for leadership decision-making Streamlit App
CI/CD Pipeline Evidence of production-grade automation mindset GitHub Actions
Evidence Pack Sample executive-ready outputs I generate for leadership reviews docs/evidence/
Program Artifacts RAID logs, decision logs, exec updates — showing operational rigor docs/templates/

Dashboard Preview

Dashboard preview

(High-res backup: docs/images/dashboard.pdf)

💼 What This Demonstrates (Using Synthetic Data)

Business Challenge How I Solved It Result
CapEx variance blind spots Automated variance tracking by program/category/month with root-cause tagging +$7.5M variance surfaced early across $561.8M plan
Readiness status ambiguity RAG-scored readiness gates with dependency-aware critical path 57.5% → 87.0% readiness clarity across 50 tools
Expedite cost leakage Vendor-level burn analysis with driver categorization $7.6M expedite tracked across 1,434 lines
Leadership reporting overhead CI-generated evidence packs on every commit Zero-touch exec-ready outputs

Dataset scale: 5 programs, 50 tools, 6 categories, 6 vendors, 24 months — all synthetic CSVs in data/raw/

🏗️ Architecture & Design Decisions

┌─────────────────────────────────────────────────────────────┐
│  Leadership Layer (GitHub Pages / Markdown Evidence Packs)  │
├─────────────────────────────────────────────────────────────┤
│  Analytics Engine (Pandas + Plotly + Custom Logic)          │
│  ├── Readiness scoring with dependency-aware critical path  │
│  ├── CapEx variance analysis with forecast drift detection  │
│  └── Expedite burn-down by vendor & root cause              │
├─────────────────────────────────────────────────────────────┤
│  Data Layer (Synthetic CSVs → Extensible to ERP/PLM APIs)   │
└─────────────────────────────────────────────────────────────┘

Key Design Choices:

  • Synthetic data only: Demonstrates capability without exposing proprietary information
  • Modular analytics: Each module (readiness.py, critical_path.py, expedite.py) reusable across programs
  • CI-generated outputs: Mirrors production automation of leadership reporting

✅ TPM/OPM Competencies Demonstrated

Competency Evidence in This Repo
Cross-functional orchestration Integration of facilities, supply chain, and finance data models
Executive communication Automated evidence packs + RAID/decision log templates
Financial acumen CapEx variance analysis, forecast drift, expedite ROI tracking
Risk management Critical path analysis, gate slip risk scoring, RAG statusing
Process automation CI/CD pipeline for zero-touch reporting
Data-driven decision making Plotly dashboards with drill-down capability
NPI/Operational excellence Tool readiness gating, install → power-on → SAT tracking

What Questions the Dashboard Answers

  • What's on the critical path right now (per program/tool)?
  • What's blocking install → power-on → commissioning → SAT?
  • Where are we burning expedite, and which vendors drive it?
  • Which gates are most likely to slip, and why?
  • Where is CapEx trending vs plan/forecast (what's driving variance)?

What's Included

1) Streamlit Dashboard

  • Entry point: app.py
  • Reads from: data/raw/ (synthetic CSVs)

2) Analytics Modules (Reusable Program Logic)

  • src/analytics/readiness.py — readiness rollups + RAG
  • src/analytics/critical_path.py — dependency-aware critical path per tool/program
  • src/analytics/expedite.py — vendor burn summaries

3) Evidence Pack (Auto-Generated CI Artifact)

Generated by: python -m src.tooling.generate_evidence

Outputs to docs/evidence/:

  • readiness_score_output.md
  • critical_path_output.md
  • expedite_summary_output.md
  • capex_variance_snapshot.md
  • gate_slip_risk_output.md

How to Run Locally

Prerequisites: Python 3.11+

# Setup
python -m venv .venv
source .venv/bin/activate  # Windows: .\.venv\Scripts\activate
pip install -r requirements.txt

# Run dashboard
streamlit run app.py

# Generate evidence pack
python -m src.tooling.generate_evidence

CI / Automation

Workflow: .github/workflows/capex_readiness_ci.yml

  • Installs dependencies
  • Runs python -m src.tooling.generate_evidence
  • Uploads docs/evidence/** as CI artifact

🔒 Adapting to Production (Data Governance)

This repository uses synthetic/anonymized data only. In production environments, I implement:

  • Data classification: CapEx data tagged by sensitivity level
  • Anonymization pipelines: Automated PII/vendor identifier scrubbing
  • API integration: Direct connections to ERP (SAP/Oracle) and PLM systems
  • Access controls: Role-based permissions for program/finance/executive views

Never commit proprietary data. This portfolio demonstrates the logic — the data layer is swappable.

🚀 Roadmap (Production Hardening)

Priority Enhancement Business Value
P0 Scenario planning module (Forecast/Commit/Stretch) Enable "what-if" analysis for CapEx reallocation
P1 Automated gate go/no-go criteria Reduce program review prep from days to hours
P2 KPI suite (OTD, lead time P95, expedite rate) Standardize vendor performance scorecards
P3 Schema validation + data quality checks Prevent garbage-in-garbage-out in automated pipelines

🛠️ Tech Stack

Data & Analytics: Python · Pandas · NumPy · Plotly
App & Visualization: Streamlit · HTML/CSS
Automation & DevOps: GitHub Actions · Bash
Data Engineering: SQL (PostgreSQL-compatible) · Docker-ready

Program Management Artifacts

Templates

  • docs/templates/DECISION_LOG_TEMPLATE.md
  • docs/templates/RAID_LOG_TEMPLATE.md
  • docs/templates/WEEKLY_EXEC_UPDATE_TEMPLATE.md

Samples

  • docs/samples/DECISION_LOG_SAMPLE.md
  • docs/samples/RAID_LOG_SAMPLE.md
  • docs/samples/WEEKLY_EXEC_UPDATE_2026-01-02.md

System View

  • docs/diagrams/system_view.md

Repo Structure

data/
  raw/                       # synthetic/anonymized source data
  processed/                 # rollups used by charts 
docs/
  data_dictionary/           # column-level documentation
  diagrams/                  # system views
  evidence/                  # auto-generated outputs
  images/                    # screenshots / preview PDF
  samples/                   # program artifacts
  templates/                 # program templates
src/
  analytics/                 # readiness, critical path, expedite logic
  tooling/                   # evidence generation scripts
  utils/                     # IO helpers
app.py                       # Streamlit dashboard
.github/                     # CI workflow

🤝 Contributing

This is a demonstration project for portfolio purposes. To extend:

  1. Fork the repository
  2. Create a feature branch
  3. Add enhancements (new models, visualizations, data sources)
  4. Submit a pull request

📬 Connect

Sourabh Tarodekar | CapEx Program Management · NPI Operations · Portfolio Analytics

LinkedIn · Email · Full Portfolio

📄 License

MIT License — See LICENSE file for details