EXPLORE OSOL

Case study · Supply chain

From blind spots to verified deliveries

VeriShipment is an AI-powered verification system built in-house by Fatima Group IT for Fatima Fertilizer. It confirms that the same truck and load that leave a warehouse arrive at their destination.

  • Computer vision
  • Logistics
  • Shipment verification
  • Supply chain integrity
Forklift driver lifting a pallet of canned goods onto a conveyor in a warehouse

At a glance

VeriShipment, in four facts

Product
VeriShipment
Industry
Logistics and supply chain
Built for
Fatima Fertilizer warehouses
Outcome
PKR 7.5B+ in freight secured
Same truck. Same load. Proven at both ends, across 2,000+ verified transactions.

VeriShipment · The design goal

The case file

From challenge to outcome.

Step 01

The challenge

Uploaded videos were reviewed by hand, deliveries were recorded without proof of arrival, and gaps between dispatch and receipt skewed stock and forecasts.

Forklift driver lifting a pallet of canned goods onto a conveyor in a warehouse

Step 02

Our approach

A two-score verification engine. Truck and stock are photographed in the existing mobile app, YOLO v11 isolates the truck, and a multimodal LLM and vector-similarity search score the match independently.

Three colleagues looking at a laptop together in an assembly workshop

Step 03

The solution

Verification built into the ERP dispatch and receiving workflow. Shipments above the threshold are verified automatically; low-confidence matches go to a person, and every decision is logged.

Control-room screen with a glowing circular data display and status panels

Step 04

From manual review to verified delivery

  • 8x fasterShipment verification vs. manual review.
  • 85% accurateIn live warehouse conditions.
  • PKR 8–10Cost per verified transaction.
Forklift driver lifting a pallet of canned goods onto a conveyor in a warehouse

FAQ

Questions about VeriShipment.

The short answers, in the order teams usually ask them.

What is VeriShipment?

An AI system that checks the truck and load arriving at a destination are the same ones that left the warehouse.

Does it need new cameras or hardware?

No. Photos are taken in the existing mobile app at dispatch and receiving.

How does it decide a match?

Computer vision isolates the truck, then a multimodal LLM and a vector-similarity search each score the match independently.

What happens when the AI isn’t sure?

Low-confidence matches go to a person, who makes the final call.

Can staff skip the check?

No. It’s a mandatory step inside the ERP workflow, and every decision is logged.

What has it delivered so far?

PKR 7.5B+ in freight secured across 2,000+ verified transactions, with 40% workforce hours saved.

Pilot VeriShipment on
one dispatch route.

Bring the challenge. Bring the ambition.
We’ll show you what it looks like on your own data.

LET’S MAKE INTELLIGENCE USEFUL

A real challenge. A new possibility.

Tell us what you’re working on and the Osol team will get back to you.