Machine Learning in Supply Chain: KoiReader Technologies & NVIDIA Team Up To Simplify Distribution Process
Pepsico, the global leader in convenient foods and beverages is excited to implement advanced machine vision technology from startup KoiReader Technologies, an NVIDIA Metropolis partner, to scale up the accuracy of its distribution process.
Pepsico believes that KoiReader’s technology is the ideal solution that drastically enhances the process of reading warehouse labels. The artificial intelligence-powered innovation aids in the reading of warehouse labels and barcodes in fast-paced scenarios where the labels can be any size, at any angle, or even partially obstructed or damaged.
This technology has been successfully implemented at the PepsiCo distribution center in the Dallas-Fort Worth area. The brand is planning on a broader deployment this year. Bellon, senior director of digital supply chain at PepsiCo states,
“If you find the right lever, you could dramatically improve our throughput.” KoiReader’s technology is being used to train and run the deep learning algorithms that power PepsiCo’s AI label and barcode scanning system.
When Pepsico realized that almost-perfect accuracy was achieved, it decided to expand the application to verify customer deliveries in order to guarantee complete accuracy of processes involving human-assisted picking.
Accurate Scans & Real-Time Insights
Koi’s AutonomousOCR, the technology which is used at the Dallas facility, is able to quickly and accurately scan even some of the most complex warehouse labels on fast-moving conveyor belts.
The same technology is used to help warehouse workers to scan pallets of soda and snacks. The technology is also being used to automate yard tasks as tractors and trucks enter and leave PepsiCo’s Texas distribution facility.
“KoiReader’s capability offers up the potential for many use cases —starting small and demonstrating capability is key to success,” Bellon says.
Bellon also added that currently, the system is already generating valuable real-time insights, Bellon reports.
Koi’s superior technology is simplifying the process for workers by accurately tracking even those products that are regular or irregularly shaped, with or without labels.
It also assists in:
- Calculating the amount of time it takes for the workers to pack boxes.
- Counting the items they are packing.
- The time spent on how to retrieve items for boxes.
AutonomousOCR can be treated as a real-time industrial engineering study that can answer several questions about the influence of people, processes, and technology on throughput.
KoiReader uses a wide variety of the NVIDIA Metropolis stack throughout its unique product offering and client operations.
For instance, NVIDIA TAO Toolkit, DALI, and Nsight Systems are being used to train and optimize models on large NVIDIA A6000 GPU-powered servers.
In order to offer real-time results on edge nodes powered by NVIDIA A5000 GPUs and NVIDIA Jetson AGX Orin module-enabled servers for larger-scale deployments, the NVIDIA DeepStream SDK, TensorRT, and Triton Inference Server are employed.
For an effortless process, each aspect of Koi’s applications is built cloud-native, using Kubernetes, microservices, and containerization.
NVIDIA AI Enterprise software suite’s primary goal is to help PepsiCo confidently scale up and seamlessly manage its applications and AI deployments.
“The KoiVision Platform was built to deliver logistics, supply chain, and industrial automation for enterprise customers. Our solution suite is helping PepsiCo improve operational efficiency and accuracy in its distribution process,” said Ashutosh Prasad, founder, and CEO of KoiReader.
“We’re testing out an object- and activity-detection capabilities and computer vision today to figure out what kind of data we want to collect with that sort of application,” Bellon said.
[To share your insights with us, please write to firstname.lastname@example.org].