ATLANTA – GreyOrange, which develops a range of autonomous mobile robotics systems and artificial intelligence for fulfillment operations, today announced the latest version of its fulfillment operating system, known as GreyMatter.
The company said the latest release includes features that optimize fulfillment for companies with omnichannel and e-commerce needs, as well as enabling efficient store replenishment strategies based on how individual stores prefer inventory packed, with the goal of reducing time and labor in moving stock from receipt to shelf. The new software updates also expands on the orchestration of data and actions across the GreyOrange range of mobile robots, rebranded as the GreyOrange Ranger series.
The robots use machine learning to adjust decisions and behavior based on real-time observations, GreyOrange said. Communication among the robots and the GreyMatter central system incorporates the learning so that the entire system “continues to get smarter.”
GreyOrange’s robots include:
- Ranger GTP, formerly known as Butler. The goods-to-person mobile robots can transport inventory from 220 to 3,500 pounds to workers for picking and packing.
- Ranger Mobile Sorter, formerly known as Flexo. These mobile sortation robots operate in fleets to move parcels from receiving through dispatch to avoid sortation bottlenecks that can occur in rigid systems, especially during peak volumes.
- Ranger Picking, formerly called PickPal. This is a picking robot designed to work in tandem with goods-to-person robots to either assist humans with picking orders, or to pick orders autonomously in manned or unmanned warehouses.
GreyOrange, an RBR50 2019 honoree, said the GreyMatter intelligence is integrated as a learning layer in the robots, so they can adapt to what is happening within a distribution center, as order patterns and fulfillment expectations fluctuate. The software can continuously recalculate order fulfillment priorities and inventory movement patterns based on real-time factors, such as order fulfillment commitments, actual fulfillment speeds, available resources, and time remaining in dispatch windows.
“Synthesizing GreyMatter with a family of individually purposed robots that are built to be collaborative with each other and with the GreyMatter hub makes GreyOrange unique in the industry,” said Akash Gupta, CTO at GreyOrange. “We’ve seen the performance benefits of designing artificial intelligence-driven software and mobile robotics together, so that each enhances the learning and adaptation of the other, rather than alternative approaches that simply interface software and robots.”
Samay Kohli, CEO at GreyOrange, said many companies continue to struggle with customer expectations for same-day and next-day delivery, along with store replenishment runs that are required two to three times per week, or even every day. “They are trying to meet modern fulfillment demands using software and hardware built for a time before Amazon changed the game for everyone by accelerating collective expectations,” said Kohli. “The idea that software and robots built together using the same intelligence is required in a modern Fulfillment Operating System is unique to GreyOrange, and makes GreyMatter the only modern solution built specifically to address modern fulfillment challenges.”
The company said its robots and software are designed for fast-paced, high-volume, and high-product-variety operations, unlike hardware and software systems that were built in an earlier time and “interfaced together in a complex technology system.” Many of the new updates are centered around data pattern intelligence, giving the mobile robots the ability to operate at scale, including doing things such as retrieving and placing inventory racks on multiple floors or mezzanines using elevator access.
The company also discussed its concept of “High-Yield Fulfillment,” which ensures that the right inventory is at the right place at the right time in order to prioritize the order of fulfillment. GreyOrange said this means payoffs from orders filled and dispatched “are greater than the tradeoffs from orders that could have been fulfilled earlier, but were assigned a later fulfillment time by the system.” These types of decisions are impacted through predictions and in real time, through data points such as inventory positions, orders, promise dates, cost impacts, revenue implications, labor available, time available, and fleets of robots available.