Robots And The Autonomous Supply Chain

Autonomous technology continues to make an impact on the supply chain. The autonomous supply chain, applies to moving goods without human intervention (to some degree at least) or aiding in achieving inventory accuracy. One of the more interesting examples is the Belgian brewery De Halve Maan, which in an effort to reduce congestion on the city streets, built a beer pipeline under the streets. The pipeline is capable of carrying 1,500 gallons of beer an hour at 12 mph to a bottling facility two miles away.

Autonomous technology is seen in warehouses and stores, on highways and in mines, and in last mile deliveries. A new entrant to the market could be autonomous freight trains, as tests are underway. Each of these areas are at different stages of the maturity curve; warehouses have fully deployed autonomous robots, while autonomous trucks are still in the early testing stages. Let’s take a look at how these technologies are being used and what the future holds.

Autonomous Mobile Robots and Drones

Today’s warehouse is becoming more automated on a near daily basis. Autonomous mobile robots (AMRs) are now commonplace in many warehouses, helping warehouse workers to fulfill orders quickly and efficiently. There are a few different types of robots that companies are considering, and each has its own unique set advantages. The key for AMRs is that they enable workers to be more productive due to constant collaboration. This is especially true as warehouse labor is becoming harder to find and more expensive to train. The autonomous mobile robot market is exploding. Ecommerce is increasing during this pandemic, so this market is likely to continue to grow despite the recession.

There are two types of AMRs – those based on fleet management and systems that rely on picking optimization:

Fleet management solutions typically operate with bigger payloads and route the robots from an origin to a destination.

Pick optimization robots integrate the movement of machines and people in a process flow designed to increase picking throughput. Pick optimization robots support picking to cartons and totes and consequently have a small payload.

AMRs are not the only autonomous vehicle technology being deployed in warehouses. Another example is the use of drones in the warehouse. When people think of drones, they typically envision ordering a package online and having it lowered down to their front door by the drone. However, given the stringent FAA regulations around drone deliveries, this is not quite a reality yet. Although, more and more tests are being conducted as we speak. Instead, drones are being used in warehouses and yards for inventory management. Using a combination of computer vision technology, artificial intelligence, and RFID sensors, drones are able to perform inventory management tasks within the warehouse or yard faster and more accurately than the human eye.

Store Robots that Aid in Replenishment

There are different players in this space, but Bossa Nova Robots has gotten the most attention as Walmart announced that they will be using these robots in hundreds of their stores. A previous generation of these AMRs also had cameras designed to track on-shelf inventory in stores. They did it badly. They could not really count the number of units of a stock keeping unit in a slot because they could not see the units behind the item in front. They also had a hard time identifying products that were twisted on the shelf. That is not what the Bosa Nova robots were designed to do. They identify whether a slot is empty or full. They identify which stock keeping unit (SKU) is in a slot and the price associated with that slot.

This is not trivial. Having items incorrectly priced is illegal if the goods have been promoted at a lower price in advertisements or online. It is also poor customer service. Knowing that a stock keeping unit is not on the shelf allows workers to stock the item and capture sales they would otherwise lose.

Autonomous Trucks

Autonomous trucks have been a hot topic. Testing is ramping up across the country, with more autonomous truck manufacturers and states completing pilots. One of the most famous pilots occurred about three and half years ago when Uber’s self-driving truck subsidiary Otto completed a 120-mile trip on Colorado highways to transport nearly 2,000 cases of Budweiser from Fort Collins to Colorado Springs. During the trip, Uber reported that a driver stayed in the sleeper berth to monitor the delivery. While the test was great publicity, it did not mean that the world was ready for fully autonomous trucks on the highway. Within two years, Uber shuttered its autonomous truck division to focus on its Uber Freight marketplace. Other players in the autonomous market include Embark, Daimler/Mercedes, Volvo, Tesla, and TuSimple.

For a fully autonomous trip that would involve starting at an urban warehouse and ending at a customer location in a different urban area, there are still incredibly significant technological hurdles. It is likely we are still years away from that. Interestingly, regulatory hurdles are not that high in several states. However, a significant risk to the industry is that a crash involving the loss of life of some innocent family on the Interstate leads to much tougher state regulations. Public perception around the safety of these vehicles will play a big role in the future of autonomous trucks. 

Venture capital funded startups have poured money into tests. But in some cases, investment dollars are beginning to dry up for this technology. One of the biggest names in autonomous truck technology has been Starsky Robotics. It was at the forefront of putting autonomous trucks on the road. Its list of accomplishments is staggering. In 2016, it became the first street-legal vehicle to be paid to do real work without a driver behind the wheel. In 2018, it became the first street-legal truck to do a fully unmanned run. In 2019, it became the first fully unmanned truck to drive on a live highway. And now, even with these accomplishments, due to a lack of funding and perhaps unfair expectations, the company is shutting down.

The Starsky model was a remote driver in the control tower navigates the truck from a warehouse to the interstate, at that point the truck would take over and become autonomous. At the other end, a driver in a control room would again take control of the vehicle as it exited the Interstate.

It is likely that we will evolve to full autonomy in a series of steps from simpler semiautonomous models. From this perspective, a company called Locomation is attacking the problem with a platooning model that makes a lot of sense. Their model is a two-truck convoy, with an active driver in the front truck, and a passive driver in the following truck. The vehicles move as a group with the lead truck communicating braking and acceleration parameters to the following vehicle. The driver in the following truck is resting, the hours of service clock does not apply to this driver until he shifts positions with the lead driver and assumes responsibility for driving. Even this is more difficult technologically than you might think. In the early pilots has started with identical trucks, identically maintained, with similar brakes and tires, and maintenance schedules. These similarities make the task of synchronously braking exactly at exactly the same speed much easier.

There are examples of practical use cases that are ongoing. Autonomous trucks are operating at iron ore mines in Pilbara, Western Australia. The trucks were brought in to alleviate safety concerns for drivers, while increasing efficiency. These trucks are mostly operating on deserted dirt roads not open to the public.

Autonomous Last Mile Deliveries

Last mile deliveries typically are the most expensive and difficult part of the B2C supply chain. And as companies look to make deliveries more efficient, autonomous vehicles are starting to come in to play. Many companies have begun testing autonomous delivery bots in cities and on college campuses. Starship Technologies has been one of the most prominent names in autonomous delivery bots. The company has moved beyond the pilot phase of home deliveries and has launched an on-demand package delivery system. Another example is Postmates, which has dispatched delivery robots in the Los Angeles area.

Autonomous technology has improved, and the interest is growing. As a result, autonomous vehicles have also been tested for home delivery. In this situation, an autonomous vehicle brings a package to a customer’s house, whereupon the customer enters a code and retrieves the package from a cargo hold. Ford has run multiple pilots to gage customer perception and acceptance of autonomous vehicles as a delivery mechanism.

Autonomous Freight Trains

Autonomous freight train testing has occurred in Colorado. In the test, a 30-car freight train led by three diesel locomotives covered 48 miles through the Colorado desert. The demonstration at the Transportation Technology Center was the debut of driverless train software. Federal regulators are beginning to clear the way for autonomous freight trains and have indicated that anti-collision technology could prevent fatal accidents from occurring. Many rail companies have begun implementing artificial intelligence and machine vision systems as part of safety protocols, which could be the beginning of a transition towards autonomous trains.

The interesting thing for freight trains is that there are currently no US regulations preventing autonomous trains from operating. Last year, the Federal Railroad Administration (FRA) withdrew a proposed rule that would have required two people on all freight trains because it would “stifle innovation.” Even so, public safety and perception is clearly the biggest roadblock for freight trains. Unlike some other autonomous trains around the world, like Vancouver’s Sky Train and Copenhagen’s Metro Loop, freight trains operate in highly trafficked areas, which poses a concern for public safety. The only operating autonomous freight trains in the world are at mines in Australia, which as mentioned above, are off the beaten path and generally do not encounter public traffic. But, as more companies look to implement artificial intelligence and machine vision, the push could be there to move to move to fully autonomous freight trains.

Final Thought

The autonomous supply chain is alive and well. Warehouses are deploying autonomous mobile robots and drones for efficiencies in picking orders and inventory management. Highway tests continue for autonomous trucks and opportunities abound for platooning technology. Last mile deliveries have seen autonomous delivery bots and cars operating in cities and across college campuses. And Australian mines are ahead of the curve, using fully autonomous trucks and freight trains to move iron ore. There are still a lot of regulatory hurdles that need to be cleared for the supply chain to become more reliant on autonomous vehicles, but the framework and interest is there. This will certainly be an interesting.

One of the biggest concerns about AMRs is how warehouse workers will interact with the bots. The fear is that the bots will get in the way of their human counterparts, cause slowdowns, or worse, cause dangerous working conditions. Generally speaking, from all of the companies we have spoken with, within a week or two of deployment, employees were used to seeing the bots scurrying around the warehouse and were no longer concerned about collisions.

Article by Steve Banker. Chris Cunnane, a Research Director for Supply Chain Management at the ARC Advisory Group, was coauthor for this article. To see the original article please click here.