The Internet of Things (IoT) and Artificial Intelligence (AI) are cornerstones of many Industry 4.0 business strategies, enabling organizations to collect and process data at scale to intelligently optimize systems and processes. However, in the rush to capture “Big Data” from distributed devices, many businesses overlook an essential principle: success is not simply about amassing the most data, but about finding the right data and efficiently transforming those data into actionable insights.
At Michigan State University’s DeepTech Lab, we encourage our academic and corporate partners to adopt a strategy we call “Pareto Data.” The Pareto Principle—often summarized as 80% of outcomes resulting from 20% of inputs—applies to many data-driven business strategies. The key to a successful data-driven Digital Transformation lies in identifying the minimum viable data required to meet business goals, requiring leaders to focus on efficiently capturing the 20% of input data that drives the bulk of business value. This approach helps companies turn their data into practical, actionable insights without unnecessary complexity or cost.
These principles aren’t just theoretical—they have real-world applications that can transform business operations. Take the example of a manufacturing facility. Instead of relying on a few costly high-fidelity, fast data-rate sensors instrumenting a small number of machines, a better approach might be to use a larger number of slower, less precise, and less costly sensors across an entire production line. This combination provides leadership with a broader perspective on the factory’s overall performance, uncovering trends and inefficiencies that even a large volume of data from few, high-precision sensors might miss. By focusing on the right blend of data sources and optimizing for both quantity and quality, leaders can discover directional insights that facilitate better resource allocation decisions and improve overall efficiency.
A prime example of this is predictive maintenance. By analyzing IoT sensor data, AI models can predict when equipment is likely to fail, allowing businesses to schedule maintenance before costly breakdowns and downtime occur. Instead of being swamped by a deluge irrelevant data, companies can focus their capture strategy to easily zero in on key indicators of equipment health, extending the lifespan of assets and minimizing interruptions.
At Michigan State, we’ve successfully applied this strategy across industries. In a project involving a fleet of vehicles, we used inexpensive sensors to collect acoustic data from vehicle engines for predictive maintenance. Rather than relying on traditionally poor-performing automotive onboard sensors, or dedicated, high precision powertrain monitoring devices, we deployed low-cost mobile phones to capture audio from an entire fleet. This approach yielded a reliable, if coarse, summary of the fleet’s health. These insights helped the fleet manager identify inefficiencies, schedule maintenance, and improve productivity—all while reducing costs. Transportation is just one example; this strategy has applications across industries from manufacturing and logistics to healthcare and energy. By focusing on “Pareto Data,” businesses can optimize performance and reduce costs, regardless of the industry.
For business leaders, the message is clear: the value of data lies not (solely) in their volume but also in their relevance. Thoughtfully blending data sources and collecting only those data that matter to operations within a businesses’ control, leadership can unlock hidden insights that help to reduce complexity, cut costs, and improve efficiency. Finding the Pareto-optimal “sweet spot” of data capture can yield impressive returns on data and infrastructure investment, often unlocking 10-20% improvements over “business as usual” solutions while retaining a fixed digitization budget.
IoT and AI are powerful tools for operational transformation, but their effectiveness depends on identifying and applying the right data. In Industry 4.0, aligning your data strategy with your organization’s strengths and needs is essential to maintaining competitive advantage. By adopting a targeted, Pareto-optimal approach, businesses can harness the full potential of IoT and AI technologies, turning small, well-chosen data points into powerful business insights without requiring a full Big Data buildout.
