Within manufacturing, energy spend is a line item that can stealthily gnaw at the margins if it’s not measured and responded to in real-time. India Glycols had a need for visibility in power consumption at a major manufacturing plant in Uttarakhand. The power spend was a hefty one on the cost sheet, but the system would not receive energy information timeously enough to facilitate immediate responses.
The issue was not simply about consuming too much power. It was also about the lack of timely, reliable information on how much energy was being drawn from the grid, generated in-house, or procured through exchange-based arrangements. “What we were seeing is that entries were getting made in SAP much later, sometimes after days or even weeks. That left very little opportunity to identify where we could optimize energy spend,” Atul Govil, Chief Transformation Officer & Head (SAP & IT), India Glycols, says.
For a manufacturing business, that delay has real consequences. When energy data is updated late, batch costing becomes less accurate, deviations from normal consumption patterns go unnoticed, and pricing decisions can be affected. “If the production cost itself changes significantly because of energy and other input costs, profitability takes a serious hit,” he says.
Leveraging Smart Energy Meters, Analytics and Cloud
To address this, the team implemented 120-130 smart energy meters across critical panels and equipment, creating a near real-time monitoring system powered by analytics and cloud capabilities. The goal was not only to capture consumption data, but also to match it against generation and understand the balance across the plant as electricity can’t be stored.
The project also helped uncover deeper operational inefficiencies. In some cases, equipment had been sized for older production requirements and was now running at suboptimal load levels. In others, the data pointed to issues that were previously invisible, including process gaps and operating discipline concerns. “We found that some equipment was running far below its ideal efficiency. That was a proxy for problems that would otherwise never come under the radar,” Govil says.
The response included corrective actions such as variable frequency drives, equipment rebalancing, and in some cases, interchange of assets within the plant. This was not a simple technology deployment. It required shutdown planning, coordination across teams, and a sustained effort to build internal confidence in the new model. “We had to sell the value internally,” Govil notes. “People had been operating these plants for 20 or 30 years, so it took time to show how this would create day-to-day value.”
Once the system stabilized, the impact became clear. Manual reporting reduced significantly, teams began using dashboards and alerts to act faster, and the organization gained a single source of truth for energy-related decisions. “The focus shifted from collecting data to making sense of the data on a real-time basis,” he says.
The initiative delivered an estimated 3% to 4% optimization in energy costs. More importantly, it changed the operating model. Energy management was no longer a retrospective reporting exercise; it became part of the plant’s daily decision-making rhythm. “It was not just a tech product,” Govil says. “It was a radical shift in the way we operated.”
India Glycols is now looking to build on that foundation by introducing an AI layer to extract deeper insights from the platform. As Govil puts it, the next step is to “bridge AI” into the system so the organization can derive even more value from the data already being captured.
