Supply Chain
Analytics

Analyzed 180K+ supply chain records to identify delivery bottlenecks and uncover that delays are driven by high-volume segments rather than inefficiency.

Read Full Analysis View on GitHub
Scroll
Revenue
$12.4M
Orders
180K+
Profit Margin
28.3%
Late Deliveries
57%

Business Problem

Nearly 57% of orders were delivered later than scheduled, indicating widespread fulfillment challenges across the entire supply chain network.

The goal was to identify where delays have the highest business impact and prioritize operational improvements at the most critical leverage points.

57%
Orders delayed
180K+
Records analyzed
$12.4M
Total revenue

Key Insight

While premium shipping modes show higher relative delays, the majority of delayed orders are driven by high-volume Standard Class shipments.

A small number of high-volume segments — primarily Standard Class shipments in Europe and Asia — account for a disproportionately large share of total delays.

Targeted improvements in these segments can significantly reduce overall delays — this concentration is the opportunity.

Standard Class
82%
Second Class
54%
First Class
61%
Same Day
38%

Share of total delayed volume by shipping class

Dashboard Preview

An interactive Power BI dashboard surfaces real-time KPIs, delay breakdowns, and segment-level drill-downs across 180K+ records.

Dashboard preview

Top Bottlenecks

A contribution vs. over-index analysis pinpoints exactly which segment × region combinations demand immediate operational attention.

Bottleneck analysis

Analytical Approach

1
Star Schema Modeling
Designed a dimensional model to separate facts from dimensions — enabling fast, flexible querying across 180K rows.
2
Power BI Dashboard
Built an interactive dashboard with slicers, drill-throughs, and conditional formatting to surface hidden patterns.
3
DAX-based KPI Design
Engineered calculated measures for delivery rate, delay ratio, and revenue contribution using DAX time intelligence.
4
Contribution vs. Over-Index Analysis
Identified which segments punch above their weight in driving delays versus which are high-volume but efficient.

Key Findings

~57% of orders are delayed — indicating systemic operational issues across the network, not isolated incidents.

Standard Class drives the majority of delays by volume — even a small improvement here has outsized impact on overall performance.

First Class shows higher inefficiency but lower absolute impact — useful for premium SLA optimization separately.

Revenue follows a long-tail distribution — a small segment of customers drives a disproportionate share of total revenue.

Tools Used

Python (Pandas) Power BI DAX Data Modeling Star Schema

Ready to Explore Full Analysis?

Dive deeper into the supply chain insights and discover actionable patterns.

← Back to Projects