Robotics and Inventory Systems

When founder and CEO Jeff Bezos announced that the company was eyeing the possibility of delivering packages from warehouse to door by means of miniature drones, he wasn’t confusing aircraft for pie in the sky. While such machines are some years away from being a practical reality, the very idea of them speaks to revolutions in both the supply and fulfillment chains and in everyday robotics, with automated systems expected to become ever more prevalent in doing fulfillment work once done by humans.

Planning for the anticipated growth in e-commerce requires vendors in turn to plan for design and implementation of robotic systems in their fulfillment and inventory chains. One simple application is the use of robots to automate forklift operations, with guided vehicles that can be programmed to move and deliver pallets and carts. On a smaller scale, Kiva robots, which cost about $25,000 each and are manufactured by a company acquired by Amazon in 2012, travel around warehouses bringing smaller storage units to human workers for picking and packing.

With robotic systems come substantial savings in the cost of fulfillment. For one thing, they will significantly reduce the number of human workers who are needed on the job, although human workers will always be required—not just to operate and maintain robotic systems, but also to augment them by, say, picking and packing single-item orders that a human can fulfill more efficiently than a machine can. Jones Lang LaSalle observes, for example, that while Amazon’s 1 million-square-foot facility in Chattanooga, Tenn., is fully “robotized,” it still employs more than 2,000 people. New Amazon plants coming online around the country feature a similar mix of human and robotic workers.

Robotic systems are expensive, even more so if they sit idle during non-peak periods, which means that those systems must be designed with an eye to how they are deployed during periods of both high and low consumer demand. A business strategy that keeps orders coming in at every hour of every day is an obvious solution, but until that ideal is met, midsize companies may be slower than larger ones to automate fully. Meanwhile, planners are challenged with creating inventory management systems that can scale easily to handle peak periods. Such systems – which make use of technology such as RFID (radio frequency identification) tagging, mobile barcode scanners, browser-based inventory software and the like – are particularly useful to companies in fields such as fashion that experience marked seasonal ebbs and flows and that need to be especially attentive to just-in-time, demand-driven stock replenishment.

Software and technology spending is now the largest cost category in supply chain investments. Because of the costs associated with planning for and purchasing robotic and other types of automated systems and their inefficiency during times of low demand, it seems reasonable to assume that these systems will not be overtaking the fulfillment process completely—at least not in the near term. No one system can be applied to all products: Food vendors will need different types of apparatus from those used by fashion retailers. Existing systems also are not perfectly scalable, meaning they cannot be easily duplicated for small, medium and large applications. The implication for real estate developers is that e-commerce fulfillment centers must be tailored to their users, and that build-to-suit development will likely continue to be the preferred route taken by retailers and third-party logistics providers alike.