Information Management and IoT: The Future of Lean Manufacturing
(Originally published in AandCtoday.com)
Lean manufacturing in today’s industrial world surrounds the continual investment in technologies that create added value, while reducing waste. As an increasing number of facilities move toward this approach, the processes involved in connecting all these various, disparate systems becomes increasingly complex. This is especially true for operations running under the Industrial Internet of Things (IIoT) umbrella.
However, as a broad spectrum technological movement seeking to connect as many inanimate objects as possible for the sake of extrapolating value, it is extremely challenging to uncover the most efficient and equitable methods for moving immense amounts of data and connecting all the necessary dots. Depending on the sector, the potential benefits of automating these systems even vary greatly throughout the world. Although, the industrial arena still holds the most potential for reaping immediate IoT cost benefits.
IoT and The Gap
Among the valued opportunities presented for industry through IoT are the better ability to plan downtime, create new efficiencies, reduce labor costs and significantly raise process data rates, which can help push lean and eco-friendly initiatives even further. This includes filling the gap between the supply and demand side of manufacturing with real-time, accurate, and actionable information that positively impacts everything from logistics and raw material sourcing to the delivery of the final product. The key is turning current systems into fully automated information management environments that provide the answers and insights needed to advance decision-making to the next level.
Why IoT
Of course, real-time data is core to such systems. The ability to instantaneously access information and create clear, high-definition pictures of every piece of the operational puzzle is fundamental to success. Unfortunately, machine learning and AI are still evolving and incapable of fully automating these processes.
But, we are on the verge of a new dawn as an ever-growing number of companies continue to develop and perfect the hardware and software needed to make IoT a reality. This will be achieved through a combination of hundreds, if not thousands, of modular solutions working in unison rather than the implementation of one comprehensive, cookie-cutter methodology. Instead of upgrading to an IoT system that does everything at once, such solutions will involve individual upgrades stacked with the ongoing influx of future enhancements resulting in comprehensive and customizable IoT systems unique to each facility.
Power Demands and Consistency
Either developed through research teams committed to customized answers or the joint efforts of multiple companies, the latest solutions are moving toward the edge and implementation of components used to process collected data through predominantly wireless processes. Initially, these facility-wide IoT capabilities will be powered by batteries provided at the lowest possible entry cost.
Long-term economics, however, will demand more cost-effective alternatives. Systems that require sensing, processing and wireless communication all at the edge consume too much power for a battery with a reasonable lifespan. Even though replacement battery costs are low, the man-hours needed to replace each battery every time it runs low will be more than enough to overcome any short-term price benefits. This is especially true with the eventual installation of tens of thousands of IoT devices spread throughout an entire facility.
One immediate workaround is to operate with lower data iteration rates. But, this is not an ideal remedy. While some processes might initially function well enough with a single upload a day, a continual flow of real-time data should be the desirable objective for providing the most short- and long-run operational value.
Energy Harvesting (EH) for the Big Picture
For these reasons, many facility managers nationwide have started to look towards energy harvesting (EH) for the solution to their data management problem. Like alternative energy sources, energy harvesting is the low power version of harvesting ambient energy. There are many different EH methodologies, but currently at the lead are photovoltaic, electrodynamic, and piezoelectric. Photovoltaic harvests energy from either solar or indoor light. Electrodynamic funnels its energy from motion sources. Piezoelectric produces energy from small crystals when deformed.
At present, photovoltaic and electrodynamic are capable of the highest power outputs. But, numerous power output hurdles exist relative to environment, intermittency, and size. While an energy harvester might perform well in a controlled laboratory environment, its true performance in the field may change wildly. Additionally, the same intermittency problems that plague alternative energy sources also pose numerous energy harvesting problems. The prospect of powering grids solely with solar is exciting, but still unfeasible given the inconsistency issues related to weather and the time of day. Similarly, a small energy harvesting system may easily run into varying conditions, which limit its ability to provide consistent power. Inefficient or inconsistent light sources, changes in the amount and rate of motion, or changes in temperature difference are just a few such examples. That can affect different harvesting methods separately. However, this goes to show that a stand-alone harvesting model can only be reliable under very specific circumstances.
Customization is the Key
Early stage energy harvesting products have typically been developed with strict performance metrics designed to limit as many variables as possible. Among the early returns is a piezoelectric powered light switch, which needs no wiring and can be mounted anywhere. With a push of a button, it sends an on/off signal to a connected light bulb powered by a consistent amount of power. However, while this is a great example of an effective energy harvesting powered solution, the method does not provide an overly effective pathway to energizing hundreds of IoT system layers operating thousands of devices.
That’s why customized solutions are an absolute necessity and offer far more productive choices than methods operating under uniform energy harvesting platforms, From a manufacturing perspective, the ability to mass-produce the same platform is attractive. However, the problem is that one-fits-all approach will not consistently and constantly work with the same efficiency in different environments.
This is where research and development is crucial to next stage energy harvesting in the IoT. In order to design a reliable energy source, there are multiple pieces of information that must be known including the availability of ambient power sources; the consistency of those sources; the amount of power needed; the method of collecting and transferring data; and the ideal data rate, among others
With these factors in mind, energy harvesting solutions are now currently being customized to specific environments with the goal of overcoming power inefficiency problems. Once tested and proven, mass production will both lower the costs and risks associated with new EH systems. As these individualized systems are developed they will also pave the way for more “tried and true” options available for an ever-widening world of applications.
However, for IoT to fully impact industry, more data will have to be accumulated from millions, if not billions, of points. Real-time information is the key to reducing a company’s unnecessary waste, negating its environmental impact, and increasing the profit margin on nearly every good produced. This will be coupled with streamlined systems that continually monitor their own inefficiencies and create further enhancement opportunities driven by the unquenchable collection and analysis of real-time data.
