Fulfillment Forecast & Routing Optimizer (FFRO)

Role: Product Management & Systems Architecture
Company: Amazon (Concept Case Study)
Year: 2025

AI-assisted forecasting and routing recommendations for Amazon fulfillment and last-mile operations.

This project is a high-fidelity MVP of an internal tool designed to help operations planners visualize network risk and make proactive routing decisions.

🔗 Repository: github.com/lelandsequel/amazon-fulfillment-optimizer

---

🚩 The Problem

Planners rely on fragmented dashboards and manual workflows to identify risk and make routing decisions. This leads to reactive planning, inconsistent service levels (SLA), and suboptimal capacity utilization.

⚡ The Solution

FFRO simulates the logic of an expert logistics planner. It ingests scenarios (Origin FCs, Destination Regions, Time Windows) and uses AI to:

1. Forecast Risk: Assign a "Service Level Risk Score" (0-100) to the network and specific regions.
2. Identify Drivers: Pinpoint specific issues like weather events, labor constraints, or carrier capacity.
3. Recommend Actions: Provide strategic, prioritized routing changes to mitigate risk.

🚀 Key Impacts

• Improved SLA performance through early detection of regional stress zones.
• Faster, more consistent routing decisions across regions.
• Reduced operational cost from better utilization and capacity balancing.

🛠️ Tech Stack

• Framework: Next.js 16 (App Router)
• Language: TypeScript
• Styling: Tailwind CSS 4 (Amazon-inspired aesthetic)
• AI Engine: OpenAI GPT-4o (via Node SDK)

📦 Getting Started

1. Clone the repo:

   git clone https://github.com/lelandsequel/amazon-fulfillment-optimizer.git

2. Install dependencies:

   npm install

3. Configure Environment: Rename .env.example to .env and add your OpenAI API Key:

   OPENAI_API_KEY=sk-...

4. Run Development Server:

   npm run dev

   Open http://localhost:3000.

---

This is a conceptual case-study tool built for a product management portfolio. It is not an official Amazon product.