For years, intralogistics operations have been trapped in a mostly manual status quo, where the majority of tasks and activities performed within distribution and fulfillment centers are done by human beings. But we are now on the cusp of an explosion in automation, ignited by the increasing sophistication of robotic solutions designed for the warehouse. These “smart,” autonomous robots not only automate the “tasks of the hands”—such as picking and packing—they also share data in real time for faster and better decision-making. They can sense their environment, carry out computations to make decisions, and perform actions automatically, rapidly, and accurately.
Experts predict that the use of robotics in supply chains will rapidly go beyond fringe implementations to mainstream use and being considered “table stakes.” The analyst group Gartner says, “By 2026, 75% of large enterprises will have adopted some form of smart robots in their warehouse operations.” This rapid adoption is being driven in part by the ability of smart robots to lower touches and reduce reliance on manual labor. But there are other factors at play as well, including:
- Cost and lead times for smart robots are dropping;
- Use cases are multiplying (as can be seen in the case studies at the bottom of this article);
- Return on investment (ROI) is improving, thanks to robots that require limited or no coding and “low-touch” robotics systems; and
- The intelligent logistics software that controls robotic processes (for example, warehouse management systems, warehouse execution systems, and warehouse control systems) is maturing.
At the same time, technology enablers for robotics are converging. These enablers include:
- Sensors and vision technology that give robotics the ability to see, sense, and interact with objects;
- Standard interfaces and open protocols that enable system wide connectivity and communication between robots;
- Edge and cloud computing that facilitate high-speed data analysis and data sharing, as well as real-time operations management; and
- Machine learning and artificial intelligence (AI) that makes autonomous robotics possible and enables continuous improvement .
Everything is coming together to mark the beginning of a golden age of robotics. There is already an enormous range of robots on the market. This includes standalone robots for tasks like assembly, picking, and transportation, as well as cobots and even flying robots for inventory management. At the other end of the spectrum are comprehensive solutions that automate entire intralogistics processes. All of these robotics vary widely and have impacts on implementation timeframes, performance, and payback expectations.
In this article, we focus on autonomous robotics for core intralogistics processes, because they have the biggest potential to make a positive impact on overall operations. Today, there are autonomous robotic systems for every core logistics process, including storage and retrieval, conveying and sorting, picking, packing, and palletizing. These types of robots range from shuttle systems and picking robots to pocket sorters and autonomous mobile robots (AMRs). The sidebar below presents four proven applications of autonomous robotics in distribution and fulfillment operations.
Getting robotics right
When you think of a robot in the warehouse, the first thing that probably comes to mind is an AMR zipping around the facility transporting orders or a robotic arm picking goods. But a robot is so much more than the hardware that you see on the floor. It’s an integrated system composed of hardware, software, vision technology, sensors, and interfaces.
Almost half of the respondents (46%) to a “2022 Intralogistics Robotics Study” by Peerless Research Group recognize that fact, saying they would prefer to buy their robotic solution as an entire integrated system—a pure capital expenditure initiative—that includes hardware, software, support, and maintenance.
In fact, key to the development of this blossoming golden age of robotics is the software used to direct and guide this next generation of robots. Robots are only as good as the software that drives them. Autonomous robotics for distribution and fulfillment operations require three types of software:
1. Intelligent logistics software for dynamic orchestration of complex distribution and fulfillment processes. This software includes warehouse management systems (WMS), warehouse control systems (WCS), and systems for gathering and analyzing the real-time data coming from the robots and other machines, as well as packing software, analytics, computerized maintenance management systems, and more.
2. A universal AI platform that can be applied to any use case or customer environment. The AI enables robots to quickly learn to manipulate objects without being told what to do.
3. An automation data system for the collection, distribution, and maintenance of item attributes. It turns out, robots need a lot of data to function efficiently. For example, fully automated palletizing systems may require up to 50 item attributes for correct handling. Insufficient automation data leads to damaged items, downtime, and, in some cases, cleaning costs.
For robots, data drives performance. Unfortunately, today’s master data wasn’t designed for robotics. It often doesn’t include information like packaging type, contents, center of gravity, tilting behavior, stackability, and pickability, to name a few.
Manually recording all this data takes time and money, and the quality of the data may suffer. You have to look at the item, measure its dimensions, enter all the data, and check twice to make sure everything is correct. It’s not an efficient process.
An automation data system records all necessary article attributes in less than 30 seconds. It essentially decodes an item’s “DNA” and adds this information to the master data. The software also distributes, manages, and continuously improves data quality using self-learning AI. It even shares article properties across networks to avoid damage during transport. It provides one central source of data.
Build a business case before you buy
Although smart robots hold great potential to transform distribution and fulfillment operations, it’s important to have a plan and business case in place before trying to implement them. As the 2022 MHI Annual Industry Report says, “The number one barrier to adoption is the lack of a clear business case.” Successful implementation of autonomous robotics requires a thoughtful approach before you buy. The following steps can help you evaluate your options and choose the right robotic solution for your operation:
1. Define your near-term objectives. What are you trying to do? Alleviate labor pains? Accelerate throughput? Increase storage capacity? Improve employee and customer experiences? Do more in the same space? Enable sustainable operations?
2. Determine your key decision drivers. What will drive your decision? Cost? Lead time? Cubic utilization? Throughput? Prioritize these to ensure you make the best decision based on your business drivers and available budget.
3. Evaluate solutions and vendors. It’s important to be aware of the variety of available applications. The sidebar below presents just four of the many use cases. Furthermore, don’t just buy a robot. Buy a system and the organization that backs it. Investing in robotics shouldn’t be transactional. It should represent the beginning of a mutually beneficial and ongoing relationship.
4. Create a multi-year map for your journey. What are your long-range goals? Do you ultimately want to achieve a fully autonomous supply chain? Build resilience? Get to net-zero greenhouse gas emissions?
Evaluating and planning a robotic implementation can take time, but it is important not to delay the process. As Gartner says, “Supply chains will become autonomous faster than you expect.” Supply chain leaders are already making autonomous robotics for core intralogistics processes a key component of their strategies. You should too.
Autonomous robotics is a rapidly growing area of automation in the warehouse with an
increasing number of possible applications. The following case studies present just
four examples of autonomous robots and possible applications.
The incredible shrinking process
One of the world’s largest retailers was looking to build a next-generation fulfillment network that would be able to rapidly pick and ship online orders to meet aggressive delivery agreements. The retailer also wanted to create a positive workplace that would attract and retain employees.
The core component of their high-tech fulfillment centers is a massive robotic shuttle system that spans from floor to ceiling and maximizes every square inch of its footprint. With this advanced system, the retailer:
- Doubles storage capacity, because the system can accommodate millions of items,
- Doubles the number of orders they can fill in a day,
- Improves the comfort for employees who no longer have to walk up to nine miles each day to retrieve goods, and
- Creates new tech-focused jobs that provide more meaningful work and higher wages for associates.
The robotic shuttle system significantly streamlines the fulfillment process because it packs a lot of functionality into one system. It automatically stores stock and overstock. It picks, buffers, and sequences orders. It also supplies goods-to-person workstations and replenishes other work areas.
Instead of a manual, 12-step process, the retailer now has an automated five-step process. The shuttle system seamlessly integrates with the intelligent logistics software, making it possible for the retailer to fulfill orders in just 30 minutes from click to ship.
Robots are now in fashion
Apparel picking is notoriously difficult for robots. There’s a vast range of product sizes, shapes, textures, weights, and packaging. Items change shape when picked up. Stock keeping units (SKUs) change during seasonal rotations. It’s virtually impossible to hard-code all the variables.
A global third-party logistics provider (3PL) conquered these obstacles with a robotic system that picks into a pocket sorter. The robotic picking system has a “brain,” so no hard-coding is required. It can handle virtually any item and unstructured scenario—even transparent, reflective, and floppy polybags.
The robot brain rapidly processes visual information and identifies the optimal gripper, gripping point, and gripping speed. Then the robot arm places items onto a pocket conveyor for sorting, grouping, and routing to packing stations. When new SKUs are introduced, the AI brain infers from past experiences, learns with every grip, and shares learnings with other robots via the cloud.
Item DNA created using the automation data system provides an additional performance boost. The software automatically captures item attributes, adds them to the master data, and gives all connected robots the information they need to function more efficiently.
The automation data works in tandem with the AI brain. One example? Some apparel have no suitable suction spots, so robots can’t pick them. Without automation data, articles will be rejected by the robotic picking system and sent back to the robotic shuttle system. This negatively impacts performance. With automation data, the item will be flagged during decanting and never sent to the robotic picking system in the first place.
Elevating efficiency
A global online retailer was challenged with consistently meeting consumer demands for faster delivery. The company also wanted to utilize space more effectively to accommodate a broad and constantly changing range of goods.
The answer? A pocket sorter that uses overhead space. Everything from clothing to shoes to accessories and more are conveyed, buffered, and merged to quickly complete online orders.
The robotic pocket sorter is part of a seamlessly integrated system, which includes a robotic shuttle and robotic picking and packing systems as well as intelligent logistics software. The software controls, monitors, and optimizes all of the processes. Items are handled in the most efficient way possible because the master data has been enhanced to include item attributes.
The shuttle system delivers items to both manual and robotic picking stations, where items are automatically inducted into pockets. Goods can be dropped at any location in the warehouse—without the pockets slowing down or stopping—thanks to a patent-pending pocket mechanism, which opens and closes automatically. RFID technology keeps track of every single item.
Single items are sent to pack stations in the right sequence using intelligent matrix sortation software, which also contributes to the pocket sorter’s ability to process up to 50,000 items per hour. The fully automatic pocket sorter gives the online retailer what it needs to meet today’s service level agreements (SLAs) while at the same time being scalable enough to meet tomorrow’s needs.
Shaping the future of e-grocery
A national grocer was among the first to pilot automated micro-fulfillment centers (MFCs) to handle e-grocery orders. The grocery chain deployed its first automated MFC in 2019 and opened seven new MFCs in 2021. The chain believes e-grocery will eventually comprise 20% of its business.
To increase the speed, capacity, and accuracy of its e-grocery operations without increasing the footprint, the grocer is deploying next-generation MFC technology. The company’s MFCs use autonomous mobile robots (AMRs). These “open shuttles” roll independently within the MFCs and are completely safe for use around people.
The AMRs seamlessly integrate with the robotic shuttle systems. Goods are stored in the system and picked at goods-to-person workstations. Completed orders are transported by AMRs to a flow rack and assigned a buffer conveyor. This ensures the right orders get to the right customers at the right time.
The AMRs require very little space, because they turn on their own axis. As a result, they can work in tight spaces with narrow aisles. That makes AMRs more flexible, cost-effective, and space-efficient than automated guided vehicles (AGVs). They’re faster and easier to install too, because they require:
- No structural modifications,
- No special pathways or fixed routes,
- No additional markers,
- No induction loops in the floor, and
- No human oversight during operations.
The AMRs use virtual lines defined and managed by intelligent fleet control software, which ensures smooth and efficient traffic flow. The software also ensures the vehicles avoid humans and other obstacles as they transport totes full of groceries.
The national grocer believes automated MFCs are a key element to their future success. Their e-grocery business has become more strategically important to them. Their goal is to make e-grocery a competitive advantage. AMRs will help them do it.
