Since the introduction of Scientific Management in the 1880s by Frederick Taylor, human labor in factories and other commercial venues such as distribution centers has been scrutinized with the objective of increasing the productivity of workers. Higher worker productivity translates into making more products per hour to sell, which increases profits. That’s the general theory, and the measure of productivity has become a key definition of a country’s competitiveness. Investopedia puts it this way:
“Productivity is measured and tracked by many economists as a clue for predicting future levels of GDP (gross domestic product) growth. The productivity measure commonly reported through the media is based on the ratio of GDP to total hours worked in the economy during a measuring period; this productivity measure is produced by the Bureau of Labor Statistics four times per year.”
In other words, productivity is so important, its measure has been institutionalized by the federal government.
Until recently, productivity paid no attention to the effect of robots, which have traditionally been classified under the general term “industrial automation.” There is a growing awareness, however, that robots in the workplace, be that a factory or a distribution center, are increasing in number and importance to the sustained growth of productivity.
This is an important point to consider: The maximum limits of human productivity are a reality and are becoming ever more apparent when compared to robotic capabilities. More robots are being installed each year, and a recent global study estimated that in 2014, there were 1.5 robots for every ten human workers.
Earlier this year in a blog entitled “Robots are infiltrating the growth statistics,” The Brookings Institution commented on jointly conducted new research from Uppsala University and the London School of Economics, which indicated that robots were having a significant impact on the growth of productivity. As the Brookings’ bloggers put it:
“Overall, they conclude that the use of robots within manufacturing raised the annual growth of labor productivity and GDP by 0.36 and 0.37 percentage points, respectively. That might not seem like a lot but it represents 10% of total GDP growth and 16% of labor productivity growth. Making that even more astounding is the fact that during the study’s time period robots achieved that impact while accounting for just 2.25% of the total assets of the industries studied.”
In other words, a disproportionately positive impact by robots relative to their numbers.
Robotics have not gone unnoticed by companies focused on the supply chain, and in particular, those that operate distribution and fulfillment centers. The difference between the two is where the orders from these centers are shipped. A distribution center typically fulfills orders for stores. A Wal-Mart distribution center fulfills orders to restock its own stores. A fulfillment center is usually associated with eCommerce retailers, like Amazon, and almost exclusively put together orders that are shipped directly to the individual who shopped on Amazon’s Website.
Amazon has been the most aggressive user of robotics, buying the leading builder of fulfillment center robots, Kiva Systems. Amazon has deployed more than 15,000 robots in what is called the eighth generation fulfillment center, of which the company presently has ten. Watching Amazon robots fulfill orders is a marvel to behold, but what drives Amazon to use them is its desire to improve order fulfillment productivity. In an eighth generation fulfillment center, the traditional role of the human order filler has been dramatically transformed. The human stays stationary and the robots bring the products to her for selection and packing. Highly mobile robots bringing entire shelves of products to the selector at speeds faster than a human can walk, making the stationary human more productive in the selection and packing of the order. Put another way, the slower human has been replaced by a fleet of faster robots. The limit of the human walking speed was a significant barrier to productivity growth.
In a way, Amazon’s solution is a departure from the effort to “automate” human performance in distribution centers throughout the past twenty years. Back then most order fulfillment processing was focused on shipping products to brick and mortar stores where customers went to purchase them. When Amazon went online in 1995 the era of direct to consumer fulfillment got underway.
Twenty years ago, order fulfillment in distribution centers was paper-based. A day’s worth of orders for a company’s store would be printed on paper. Supervisors would sort the work by truck route on which a number of stores would be visited. The sorted and batched orders would be handed to selectors, who would mount their forklifts and start work. Selectors determined how best to go from location to location to grab products listed on the orders.
As competitive pressures began mounting, the “follow my own selection path” methodology that gave selectors complete control of the path they would follow and the time they would take to complete an order was recognized as inefficient. Around this time, also, a new technology was introduced to better control inventory management. This was the barcode, and with it came the requirement for the connected, smart, wireless devices necessary to scan the codes. With barcode scanning, and the evolutionary leap forward in IT systems designed to manage its complexity, order selection became more “directed” and less under the control of the human selectors. It allowed for specific, system-assigned “pick paths” to be given to selectors, and with the increased data collection capabilities of the devices and supporting software, the first big improvement in productivity measurement was attained. It became possible to measure each human selector’s performance time, compare one to many and allow for the imposition of performance standards. With performance standards came incentive programs. Effectively, the selectors were offered the carrot of incentive bonuses to beat the standard, but were heavily penalized if they consistently fell below that same standard. Order selection by scanning created the first technology-based augmentation of human productivity for distribution.
Today, the highest levels of technology-enhanced human performance in distribution centers come from the use of voice-driven selection and “pick to light” techniques. Both of these focus on minimizing the need for the human selector to think about what she is doing. Rather, the prompting of next step after next step drives the selector to continuously move as instructed. In all cases, the human selector is moving from one location to the next, down a path that has been predetermined by software algorithms and delivered by devices.
Amazon changed the dynamics of the selection process when it recognized that having a human move from location to location, stop, select, and validate products and repeat this process hundreds of times a day, was far less efficient than having a fleet of robots, operating in a “human exclusion zone,” bring entire shelves of products to the human standing in a fixed location. Amazon got rid of “the walk” in their eighth generation fulfillment center.
The increase in order fulfillment productivity has been staggering. It has also been fundamental to Amazon’s drive to provide same day delivery. If a fulfillment center can now fulfill a customer order within 15 minutes from when the “buy” button was pushed by that customer, then the ability to deliver that day with the transportation options now available, such as UberCargo becomes not just a reality, but a service commitment.
I believe we are witnessing a major paradigm shift in how productivity growth will be measured, a new paradigm which will increasingly feature significant gains from robotics and the diminishment of human-based contributions. The direct human labor offset that the shift to robotics will drive will be significant, and we will see it sooner rather than later.
Tim Lindner is senior business consultant with a software company, and a regular contributor to Connected World. He can be reached at firstname.lastname@example.org