Seth Patin, chief executive officer and founder of LogistiVIEW, describes the role that technology is playing in improving the way that warehouses respond to customer demand.
Automated decision-making in the modern-day warehouse is “the process of using software to recommend that a person take an action, or automatically taking it based on the amount of information available to it,” Patin says.
Decisions that are ripe for automation include which orders should be released and when, whom the work should be assigned to, how the work should be done, and the optimal path of orders through the warehouse. “As technology advances,” Patin says, “we’re going to see more and more of those decisions automated with software.”
Artificial intelligence and machine learning are key elements in the automation of decision-making, although both have gotten “a bad rap” because most people can’t visualize anything concrete when they come up for discussion, Patin says. Simply put, AI is the application of a computer’s ability to recognize a pattern, and machine learning determines whether that pattern generates success or failure.
Data flows into the system from multiple points, both inside and outside the warehouse — so much data, in fact, that humans can’t possibly make sense of it. Critical information includes the current state of demand and supply, which orders and tasks need to be completed, the workforce that’s available to do the job, and the natural constraints of the facility. All of that is necessary to achieve sufficient visibility to make key decisions, Patin says.
The system learns with experience, although it can be effective from the very beginning because its algorithms are derived from a rules-based environment. When it confirms that a customer got an order on time and at the best cost, “the next time it sees that same situation, it’s going to reinforce that decision and do it again,” Patin says.
