Thanks to the development and utilization of new technologies like big data and analytics, business intelligence has grown by leaps and bounds in the last decade. However, that is just the tip of the iceberg. As machine learning continues to evolve due to lower costs of connectivity and sensor technology and gain widespread adoption, business intelligence is now smarter than ever and enabling truly brilliant factories.
Machine learning technology has been around for a while but it has not truly fulfilled the promise it offers. Only recently has it begun to show signs of that as manufacturers began expanding the use of machine learning from individual machines and plants to the entire network. This network-based machine learning is vitally important because no two factories are the same, but they all share the same goal: optimal performance through smart resource management. Ultimately it’s about better decision making in real time to improve the business outcome.
By enabling network-based machine learning, manufacturers can create a digital thread that weaves through every aspect of the manufacturing process — from operations to product design to supply chain — providing the type of business intelligence that is the key driver of truly brilliant factories.
Network-based machine learning algorithms will drive business intelligence further than ever before by including data sets like inventory, material cost and labor costs into a company-wide formula revealing fleet cost and material efficiencies that were previously hidden within the network of individual suppliers and plants.
Smarter, Not Harder
More important than simply increasing efficiency, with access to advanced business intelligence, manufacturers are increasingly changing the way they do business down to the most basic level. For example, by using advanced analytics and machine learning to predict maintenance issues before they occur, manufacturers can move away from the antiquated shift-focused work schedule and allow their workers to participate in the “gig economy.” Manufacturers are moving from reactive asset maintenance to condition-based and predictive maintenance strategies.
If a manufacturer has the ability to predict with a high degree of certainty that there will be maintenance issues on a piece of equipment beginning at 4 p.m., why have the mechanic on-site all day waiting for the issue and then having to work late? Instead, use this intelligence to both improve the work-life balance of the employee and save on labor costs by allowing the mechanic to come in when he or she is most needed, when maintenance issues are most likely. That mechanic is now working smarter and not harder. A shorter, more productive work day is a win-win for both the employee and the manufacturer.
Meeting Objectives Efficiently
Why push machines at full capacity when you don’t have to? Smarter business intelligence powered by machine learning gives plant engineers the ability to conserve machine resources by providing insight into the larger manufacturing process. For example, while they might be able to finish a work order for shoes in three days by running the machines at 100 percent capacity, this puts additional stress on the machines. If a network implementation of machine learning can tell operators that shoelaces won’t be ready for five days, why stress the machines to complete an order so that it can sit in a warehouse for two extra days waiting for shoelaces? By running the machines at 75 percent capacity, the work order can be completed on time, and the stress on the equipment is reduced.
These optimized workflow techniques can reduce the costs of maintenance issues and material storage for manufacturers simply by looking at the overall process and making smarter business decisions thanks to additional information.
More Effective Production
Historically, maximizing capacity has always been a struggle for manufacturers. Without the proper intelligence, it is a monumental task to ensure a diverse manufacturing operation is always running at full speed. There have always been redundancies and bottlenecks in the process. Network-wide machine learning gives manufacturers the necessary insight and intelligence to enable smarter, multi-modal facilities that run at full capacity.
A diverse manufacturer may produce air conditioners in Wyoming and refrigerators in Kansas. Before machine learning, these manufacturers would often create everything needed for these products within that same plant. Even common parts, like compressors or seals, might be individually manufactured at each plant. Now, better insight allows them to leverage multi-modal facilities and manufacture those common parts at one plant in Idaho and ship them to the plants in Wyoming and Kansas that need them.
Smarter business intelligence also enables manufacturers to reduce planned downtime by creating more collaboration and integration between previously unconnected plants. Even the best-planned, most diverse manufacturing operations sometimes have planned downtime. By connecting the entire network and analyzing the entire data set, manufacturers can use that planned downtime to help offload some of the work to other plants that are struggling.
What happens when a customer asks to push back a delivery date? Until recently, either the manufacturer or the customer had to find a way to store the products until the requested delivery date. Now, smart manufacturers can utilize their business-wide intelligence to schedule for purpose. While one customer wants to push back its delivery date, another wants to fast track its order at a different plant.
With a view across the entire network, a machine learning algorithm can more intelligently make decisions on production timelines, hold off on manufacturing the original order, and focus on speeding up the delivery of requested products from a different plant to meet both customers’ desired timelines, while simultaneously ensuring both plants run at optimal capacity.
Today’s brilliant factories, enabled by smarter business intelligence, are opening up a whole new world of business models for manufacturers both large and small. Intelligent scheduling, workflow and supply are common goals for every manufacturer and it is all based on developing and analyzing the best information possible. For larger manufacturers, that includes network-wide insight, down to the smallest detail in the largest plant. For small manufacturers, it means using the equipment you have to its full potential.
At the end of the day, the end goal of manufacturers is the same: to create high-quality products that meet the customers’ expectations in the most efficient way possible. To meet this objective, the key is to use every tool available to look across the entire operation and make business intelligence smarter.
This article originally appeared in the May 2016 issue of Workflow.