E.Con Special Series—Part One: Automated e-Commerce—Are We Ready?

Automation is playing a growing role in the e-commerce movement, according to speakers at NAIOP's E.CON conference. But it comes at a price.
Brian Jensen

Brian Jensen

Contributing Editor Erica Rascon presents a two-part series of reports on e-commerce based on discussions that took place as part of NAIOP’s E.Con conference in March.

Convenience is the backbone of e-commerce. As consumer expectations for same day and next day delivery continue to rise, retailers are seeking creative and cost-effective strategies for faster fulfillment and delivery. Automation is playing a growing role in these strategies, but experts are contesting its adaptability and value.

Furthering Fulfillment

When it comes to fulfillment, the ebb and flow of demand brings the practicality of automation into question.

Bryan Jensen, vice president of St. Onge Co., explained the dilemma to NAIOP E.Con attendees: “Automation does remove labor, but it also has to provide throughput; you have to hit the peak of that peak,” he said. “Average bucks will get you average automation and average throughput. You want maximum throughput? You want next day delivery on Dec. 20th? You’ve got to spend maximum dollars. Can you get automation to cover the peak’s costs effectively?”

Now and within the next five years, the answer for most companies is no.

Factory automation resized“Each robot can replace about two-thirds of staff,” Jensen observed. “Let’s take manual operations that need 300 workers average but 900 needed at peak. That’s a lot of workers to bring in temporarily.” Yet with an automated operation, there may only be a need for 200 workers, and 600 at peak. “That means 400 peak robots that I only need for three weeks, whereas if I bring in 600 workers, I hire them for 10 weeks. Start thinking about the numbers that go into those finances: If a robot is $30,000 each, that’s $12 million just for the robots before you even start running them, before you start putting all those shelves over them that cost about $900 to $1,000 per shelf unit.” Compared to the $1.2 million cost for temporary labor, “even with all of the attrition, you start to have a 10-year or 12-year payback period for robots.”

The costs don’t stop with the purchase of robotic equipment. Vastly automated distribution centers require better insulation and cooling, which means higher construction and operationing costs. In addition, the robots need maintenance and technology upgrades, both of which add variable costs.

Until the costs for machines and their maintenance decrease—or the cost of hiring and training temporary labor increases—automation’s greater distribution center efficiency will come at a high price.

Driving Delivery

Automation may be better suited for the delivery system, as several companies have already found ways to integrate automation into interstate travel and reduce costs. Yet there, too, it has been slow to catch on.

Four years ago, Google began to test convoys of electronic and hybrid unmanned vehicles. The trucks reduced fuel consumption by riding in the wake of the other vehicles in the convoy. These experiments came as the result of a growing shortage of CDL truckers and as an initiative to make transportation greener and more efficient. Years later, the program has yet to gain popularity. A lack of cooperation from the Department of Transportation poses one hindrance; investor and company distrust of the unfamiliar poses another.

Last-mile automated delivery has faced similar setbacks. CityCargo delivers product to inner-city retail locations and customers by repurposing existing infrastructure. Product is shipped from distribution centers to tram stations via electronic cars. It is then loaded and shipped via existing city tram rails. The last mile is carried out upon normal roads via unmanned electronic vehicles. CityCargo makes use of existing infrastructure and fuel-efficient unmanned vehicles to expedite delivery, reduce costs and drive a high level of hands-off movement in the final mile.  The CityCargo concept is more than a decade old, so why hasn’t it gained more traction?

“At the end of the day, the final mile costs—when weighed against conventional thinking and what people were comfortable with—gave way to them,” said Jensen.

Drones will encounter a similar fate if the public does not re-evaluate its relationship with robotics. While they re a viable alternative for last-mile delivery, few people are comfortable with the idea of drones flying through their cities and towns. And investors find them both costly and too unpredictable in their interference with humans.

There are other setbacks, as well. Drones have tight restrictions on flight zones and times. Furthermore, product size and weight variables require a fleet of drone sizes and logistical planning software that is still underdeveloped.

For now, costs and human behavior restrict automation to a limited role in e-commerce fulfillment and distribution. It is unclear how long those hindrances will prevent the industry from becoming predominantly automated.

“In the next five years, we’ll ask the question again,” said Jensen. “It’s a matter of timing. One hundred years from now, nothing will look like it does now.”

Part Two of the series, “The Trillion-Dollar Industry,” addressed last-mile delivery.