Supply Chain Performance Overview

Project overview

Built for demonstration using a fictional company, NorthStar Supply & Logistics, this system models 10,000 orders across products, customers, warehouses, and carriers (Jan 2023 – Dec 2024). The goal was to answer three questions in one connected system: what's driving demand, how reliable is delivery, and where is profit actually concentrated, across three linked report pages (Performance Overview, Carrier Reliability & Service Performance, and Customer & Product Intelligence), with drill-through to order-level service detail.

Headline context: $111.57M in total revenue and $39.11M in profit (35.05% margin) over the two-year window, against a 64.37% OTIF rate and a 6.91-day average lead time.

Methodology

Prepare in Excel

Structured the 10,000-order dataset (products, customers, warehouses, carriers) into a clean relational base ready for modeling.

Model in Power BI (DAX)

Built DAX measures for demand forecasting, OTIF (on-time-in-full), lead time, and profitability: the metrics that actually drive supply chain decisions, not just order counts.

Build a 6-month demand forecast

Layered a forward-looking forecast band onto the demand trend so the report answers "what's coming" alongside "what happened."

Structure for drill-through

Connected carrier and regional summary views to an order-level Service Detail page, so a flagged carrier or region can be investigated down to the affected orders.

Findings

My first instinct was to blame one carrier. The data didn't back that up. All four cluster tightly in the 64.35%–65.73% OTIF range (64.37% total). When everyone underperforms by roughly the same amount, that's a process problem (routing, scheduling, SLA design), not a "switch carriers" problem.
OTIF is trending up, just not cleanly. The rolling 3-month trend shows a real upward slope over two years, with a rough patch in the middle. Improvement's happening, but it's not steady.
Total profit and profit efficiency pointed to two different categories, which surprised me. Lubricants brings in the most total profit dollars, mostly on volume, but Hardware actually converts revenue to profit better. Lubricants wins on scale, Hardware wins on efficiency, and those aren't the same win.
I expected the biggest accounts to be getting quiet discounts. They weren't. The two highest-revenue customers (~$6–8M) sit at a similar margin % to everyone else in the Customer Value Quadrant.
East kept showing up as a problem from two different angles. Lowest regional revenue and among the highest late-order rates, underperforming on both fronts at once, not just one.

Recommendations

Since the underperformance is spread evenly across carriers instead of sitting with one, I'd audit the shared scheduling/routing process feeding all four rather than renegotiating or swapping anyone out. I'd put East at the top of the list for a root-cause review given it's weak on both revenue and reliability. On the product side, I'd keep Lubricants as the volume anchor but deliberately grow Hardware's share. It converts revenue to profit more efficiently, so even a small mix shift there lifts blended margin without touching pricing.

What this demonstrates

A multi-page, connected Power BI system (not a single dashboard) with drill-through depth.
Distinguishing "total profit" from "profit efficiency": two different questions that a single ranking can't answer.
Root-causing a reliability problem to the process level by checking whether it's concentrated or systemic first.