🚀 Executive Summary & Problem Statement
Business Challenge: Global supply chains face frequent delays from unpredictable weather and geopolitical tensions. Operations teams struggle to extract fast, actionable insights from complex Delta Lake tables.
Core Solution: Resilient-Sync leverages Databricks Genie Agent to translate plain-English queries into instant Spark SQL aggregations, returning automated text insights and chart distributions.
Quantifiable Impact: Delivers a 4x performance benchmark improvement, unlocking $635.24M in potential penalty savings.
Target Audience: Supply Chain Managers, Operations Executives, and Logistics Analysts needing immediate data access without writing manual SQL.
🛠️ Technical Architecture & Delta Lake Pipeline
Custom Simulation Engine: Built weather_stimulator.py
to model complex supply chain disruptions across 580,000+ orders.
Probabilistic Modeling: Utilized np.random.choice for weighted environmental impacts and np.random.seed for 100% reproducible testing.
Risk Intelligence Integration: Merged Geopolitical_data.csv to map maritime choke points, risk scores, and alternative routing costs directly into the dataset.
Gold Delta Lake Layer: Unified raw operational data into workspace.default.supply_chain_gold_table.
Automated Data Validation: Executed CTE-based SQL audits to ensure complete schema and row-count integrity across processing pipelines.
📊 Key Findings via Databricks Genie Agent
Order Scale & Fulfillment: Evaluated 581,093 orders—54.9% (319,113 orders) suffered late delivery, 17.7% arrived on-time, 23.1% shipped early, and 4.3% were canceled.
Top Geopolitical Risk: West Asia carries a maximum Risk Score of 9 and highest penalty cost of $65,000, followed by East Africa (Score 9, $30,000 penalty).
Route Penalty Anomaly: Central America shows a moderate risk score (5) but a high $45,000 penalty cost due to expensive re-routing options.
Weather Disruptions: Typhoons/Cyclones and Storm-induced Port Congestions cause severe average delays of ~5 days. Clear Sky (162k orders) and Heavy Rainfall (136k orders / 1.8 avg delay days) form the baseline.
Regional Bottlenecks: Central Africa recorded the highest late delivery rate at 57.4%, with global shipping delay variance tightly clustered between 0.45 and 0.62 days.
🔮 Strategic Value & Genie Capabilities
Financial Protection: Mitigates an estimated ~$2.8 Billion in unmitigated penalty exposure while protecting high-margin retail categories (maintaining 10.78% net margin).
Conversational AI Core: Genie Agent dynamically parses user intent, writes clean Spark SQL queries, and generates dual text/visual answers.
Complete Auditability: Users can view and inspect the auto-generated backend SQL code for full transparency and trust.
Democratizing Data: Enables non-technical domain experts to self-serve deep analytical risk reports in seconds.
Demo Video Link: https://drive.google.com/file/d/1gFJn2oMjtMmuRdEnrRw6k689R8vMtwjy/view?usp=drive_link
