Rising energy costs are driving up the cost of everything from manufactured goods to food. And given that Russia, a large supplier of natural gas and oil, shows few signs of halting its ongoing invasion of Ukraine — which has resulted in the US, UK and others restricting Russian fossil fuels — energy inflation is unlikely to subside any time soon, leading to warnings of larger economic fallout and even recession.
Manufacturers want to reduce those energy costs as much as anyone else — but given economic realities, it seems to some companies that the only response to energy inflation is to hike the prices of products and services. But there actually are other options — including figuring out ways to cut energy consumption, thus reducing the impact of energy inflation on production costs. And among the most efficient and effective ways to do that is with artificial intelligence. AI solutions already save energy in office buildings’ lighting and ventilation systems as well as data center HVAC systems; there is no reason the technology can’t also be applied to manufacturing and production.
Energy Waste Is Rampant
Reducing costs is often compared to cutting fat — and, it turns out, there is plenty of energy “fat” to cut in manufacturing processes. In 2017, industry experts estimated that US manufacturers wasted 66 percent of energy available in the production process — and that figure continues to grow. Eliminating even a portion of that waste could significantly tame energy expenses for producers — thus taming price rises.
Because waste is so prevalent, manufacturers have many opportunities to eliminate it. Industry experts suggest things such as cleaning and updating equipment, insulating systems, upgrading motors, and dozens of other actions. But undertaking all these steps within a short period of time is very difficult — and if the specific step taken does not provide the solution for the inefficiencies plaguing a production line, the company will have spent valuable resources on a solution that will not help reduce their energy bill.
AI Has the Ability to Reduce Energy Waste
AI systems can gather data on all aspects of the production process — like age of equipment, time of day a machine is most used, production logistics, types of raw materials used in production, and a thousand other factors — and analyze them to determine which specific factor is using more energy than it should be. Armed with that information, managers can more quickly and easily develop solutions that will reduce energy waste. The large number of sensors used in the production process collect reams of data on every aspect and phase involved in the process; machine learning systems could analyze that data to discover relationships between the factors involved in production, providing information that will enable managers to tweak the process in order to reduce energy costs.
Among the factors that AI systems could analyze are the temperature of a production facility, where temperatures that are too high or low could affect the efficiency of systems; detect inefficiencies in heating systems; help managers determine the proper size motors and drives for an application; effectively controlling production processes; and many more. All these elements could be factors in energy waste, and AI systems could examine how each of them interacts with the others in order to provide managers with information about how systems could be improved. A digital twin model of energy consumption in a manufacturing plant or industrial processing facility could provide operators with recommendations on how to calibrate systems in order to achieve maximal energy efficiency.
A production facility could have a dozen machines that produce the same component on different production lines, yet each of those machines could be using different amounts of electricity. AI systems could analyze the performance of these machines and provide answers on how to cut energy usage — by relocating them physically on the factory floor, singling out parts that may be causing the energy waste, or any other factor that could be responsible for energy waste.
The AI Implementation Process Has Started in Some Factories
AI systems are already being used by some manufacturers, most commonly for supply-chain management. But there are several smart factory projects where AI is being used to regulate energy usage in production. Sirio Pharma, based in Ma’anshan, China, recently opened the first section of its planned 240,000-square-meter Smart Factory, which, among other things, will use AI to regulate production systems, ensuring that they use available energy as efficiently as possible. According to the company, Sirio's “new smart factory, using IoT and AI, ensures an unmatched level of process control, with intelligent energy consumption and far greater logistical efficiencies.” Meanwhile, GE's multi-modal facility in Pune, India uses AI, vision technology, and other systems to prevent “cost overruns for high-volume, low-mix production on expensive machinery,” and “optimize labor and machine run time,” among other things.
While AI can effectively lower energy consumption, this approach is not yet used on a large scale. That’s partly because it’s new, and also because many companies worry about the cost of implementing such technology. Some lack the properly-curated data required for AI to work, while others who do collect data cannot effectively use it to optimize energy use. In some cases, the AI technology itself may require large amounts of energy to train; but this is usually made up for in the energy saved later. While programming AI systems to accomplish these goals is somewhat complicated, organizations can achieve significant results if the process is conducted properly.
Inflation is an unfortunate reality now — and the increased energy costs that are driving that inflation will continue to rise, at least for the near future. But producers are not without options. By cutting energy costs, they will be able to avoid, at least partially, passing off their rising costs to those purchasing their goods — thus avoiding the risk of losing business to cheaper rivals, or reducing the quality of their products by cutting corners and risking their reputation.
