The Industrial Internet of Things (IIoT) has become one of the most significant developments in the evolution of manufacturing operations. Sensor-equipped industrial equipment and control systems have been around for decades, but recent advances in the capabilities of analytics software coupled with drastically cheaper sensors has opened the door for manufacturers to make significant improvements in quality and productivity by adopting IIoT, without breaking the bank.
While much of the spotlight has focused on the use of predictive analytics to help avoid machine failures that result in downtime, plant operators are realizing that additional analytics capabilities, as part of a coordinated IIoT initiative, can also help them get much greater insight into how interrelationships among equipment on a line, as well as external factors, impact production yields and defect rates.
For example, in the process of mounting semiconductors on circuit boards, tiny puffs of air are used to direct the placement of the chips, which require extremely tight tolerances. The placement machines are calibrated to account for existing environmental conditions such as humidity and temperature, but something as simple as the presence of human operators generates heat profiles that can affect those conditions. Even small variations can result in placement defects. Sensors can monitor these changes in real time, and IIoT logic can allow the machine to automatically make the infinitesimal adjustments necessary to meet stringent quality specifications and reduce defect rates.
Deviation in source materials coming into a factory is another factor that can affect output. Take sheets of metal processed to form parts, for instance. Differences in lots or suppliers can cause the sheet metal thickness to vary. Slowing or stopping the line to recalibrate machinery to accommodate these variations damages productivity and can create quality issues. Using real-time data from machinery equipped with sensors able to detect these slight disparities, IIoT analytics and machine learning capabilities can evaluate the discrepancies and instruct line equipment to automatically adjust — all without human intervention or impairing production.
IIoT also brings together data from an entire plant, or multiple facilities, to expand a manufacturer’s insight into their overall production process. In this way, it becomes possible to predict and remediate quality problems before they negatively impact production. Plant managers can look at operational data collected and analyzed for patterns or changes in behavior to get a holistic view across the entire process, helping pinpoint where changes in machine states — or things like variations in source materials — are causing problems in output quality.
Understanding production costs is another area where IIoT can help. The current costs of operating equipment can be used to help forecast future production costs; for instance, analyzing the process of retooling a line between one product and another and discovering a new pattern of power consumption. For Product A, the machine performs a shorter function and uses less electricity, costing less to produce than Product B. Identifying these types of patterns can help enhance production planning. To that end, where manual forecasting approaches may come in with a margin of error around 20 percent, IIoT provides new, more comprehensive information that helps add extra layers of context, which can bring forecasts much closer to reality, with margins of error in the single digits.
As manufacturers accrue detailed insight into all aspects of their production, they can use IIoT to realize even greater strategic benefits, such as improving plant capacity. Historically, when manufacturers needed to increase production volume, they had to expand or build a new plant, requiring a large capital investment in order to generate greater revenues. IIoT can improve efficiency and allow greater use of existing facilities by enabling machines to be nimbler and rapidly adjust to changes in conditions or materials. Lines can be reconfigured faster as the system learns the steps and processes required for change. Downtime and human errors related to switchover drop, and switchover speed improves. The resulting runtime increase delivers greater production capacity and value.
A critical aspect of IIoT success is securing management buy-in and the effective collaboration of all stakeholders. Once management is on board, manufacturers can begin to achieve real value from IIoT by identifying one or two use cases to prove out the technology. At the same time, it’s important to establish beneficial relationships with IIoT partners, internal IT organizations, and the people charged with day-to-day plant operations to set the initiative up for success. The opportunities for business improvement are many, but focusing on improving productivity and reducing product defects can show real benefit at the bottom line. And, most importantly, it is achievable with an investment that makes financial sense.