For decades, industrial control systems have been generating enormous volumes of data, but in many cases that data hasn’t been fully employed to help companies reduce operating costs, improve reliability, and increase productivity—three goals that amount to the holy grail of manufacturing. Until recently, the path forward has been blocked by insufficient compute power, storage, and machine learning technologies to allow companies to harness the richness of the data they generate.
Today, all of this is changing. Thanks to the rise of the Industrial Internet of Things, dramatic advances in computing systems, and the rapid maturation of machine learning algorithms, manufacturers now have the ability to collect, store, and analyze huge amounts of data in real time to turn it into actionable information. They can use this information to enhance operations, remedy equipment issues proactively, improve plant availability, and meet countless other goals that drive toward better margins for the business.
Looking ahead, manufacturers see an even brighter future. We are at the dawn of an era of autonomous manufacturing, in which virtually all processes will be digitally based and highly automated. In this new world, a plant will receive an order via a digital data model that captures all of the specifications for the product, the machines will organize themselves to a great extent, and orders will transform directly into production information and flow into the production process.
So what does it take to achieve this vision of the autonomous manufacturing plant? A first step is to put the right foundation in place to enable plant managers to fully exploit the data generated on the manufacturing floor. That foundation includes a robust digital platform that captures, stores, and analyzes the data generated by manufacturing control systems and sensors on equipment connect via the IoT.
This is where the Siemens cloud-based, open IoT ecosystem, MindSphere, comes into play. MindSphere provides a platform for recording and analyzing large volumes of production data. It serves as an operating system for the Internet of Things, connecting actual production with the virtual world.
Here’s how the platform works: All data defined by the user is captured and transferred to MindSphere for analysis. Data that is relevant for optimization is made available to managers in the form of recommendations for action. This “smart data” gives managers the actionable insights they need to improve equipment uptime, increase the efficiency of production operations, and more fully utilize the potential of the plant.
MindSphere is designed to serve as the centerpiece of an open ecosystem, making it possible to securely exchange data across company boundaries and to connect a wide range of different products regardless of manufacturer in a single platform. Thanks to open standards and interfaces, data can be gathered from industrial equipment of many different manufacturers and analyzed in MindSphere.
MindSphere is currently riding a wave of momentum for automated manufacturing. In the last 12 months we have seen tremendous interest in the platform from customers in various verticals, and by now have many connected customers, ranging from machine tool OEMs and users to automotive, pharma, and other fleet-type operations.
With this game-changing technology platform in place, plant managers have a key component of the foundation they need to enable the automation of manufacturing, from the edge to the cloud. Building on this foundation, manufacturers can move from a focus on steady incremental improvements to step changes that drive a more competitive business model through end-to-end automation of manufacturing processes.
To look at the bigger picture, these are exciting times for manufacturers. They now have access to the compute power, storage capacity, networking bandwidth, and analytics capabilities needed to exploit the richness of the data they generate—and take their operations to an all-new level. The key is to understand the promise of autonomous manufacturing, and then purse the vision in a systemic manner.
For a broader look at the topics explored here, you can catch a replay of the Autonomous World Panel from the Intel 2016 AI Day. In this panel discussion, I joined three colleagues from industry and academia to talk about the opportunities and challenges of the autonomous world that we are entering. And to explore other use cases for an autonomous world, you can visit the Intel AI site.