As of 2019, 70% of manufacturing companies reported piloting or deploying new technology to digitize their manufacturing process and get on board with Industry 4.0. Successfully implementing digital supply chains has demonstrated increased efficiency and earnings, with increased productivity and reduced waste.
How are they achieving this, and what are the keys to a successful digital roadmap?
Examining factories with the greatest ROI in their digitization process reveals that successful digital roadmaps share six common features, encompassing a number of technological areas.
Six pillars of successful digital roadmaps
1: Factories adopt streaming architecture
A successful digital roadmap prioritizes moving data where it can be used. Streaming means that data is flowing from System A to System B, or from each machine to a variety of systems. Idle data sitting in a static database is useless. Upgrading the flow of data so that it travels in real-time to a variety of systems is crucial to making data-driven decisions that reduce waste and production costs. With real-time data from the factory floor shared across manufacturing systems, organizations can respond quickly and comprehensively to issues that arise. Inefficient and costly machine errors are eliminated days, weeks, or years sooner.
2: Systems collect more comprehensive and complex data
Increased connectivity and better data storage capabilities are changing the face of manufacturing data collection. Previously, manufacturing organizations might have settled for a subset of machines gathering data for a simple KPI, or one or two data fields with pass/fail information. While this data revealed important insights for the time, it left blind spots and didn’t provide the nuance and global view of operations that is possible today.
Successful digital roadmaps show factories gathering entire rich event data for what happens with a machine – and doing it for every machine across the factory floor. By bringing more complex data together from all machines, downstream users gain a complete picture not just of where machines are failing, but of how and why failures occur. They can better understand how machines work together. This rich data is powering insightful decisions with dramatic returns.
3: Multiple factory floor systems receive and utilize streamed data
With a more complex web of systems and data in place, the traditional manufacturing execution system (MES) isn’t the only destination for data. Connectivity and real-time streaming play a crucial role in this. Factories make the most of full connectivity and real-time streaming by implementing specialized systems for processing different aspects of data. Ensuring that rich, specialized data is flowing to the multiple parallel systems running on the shopfloor is a key step in taking full advantage of advanced digital capabilities.
4: Operating with advanced shop floor control use cases
Accessing predictive analytics and automation are pivotal in reducing waste as part of a digital transformation. In the past, machine A and machine B talked to each other in a one-to-one relationship to ensure agreement about isolated details, like the name of a product.
Now shopfloor control includes things like AI assist on inspection reports and is empowered to stop the line based on predictive yield data. Instead of waiting for data on the results of a low-yield process, advanced shop floor control use cases can interrupt the process to allow for a fix before a costly, waste-heavy process goes through.
By analyzing event-based data layers alongside MES data, manufacturers can access and act on complex use cases.
5: Specialized databases provide optimal solutions
Factories are finding that their new levels of data complexity and richness don’t fit well into a traditional SQL database. With data flowing to multiple factory systems, organizing data into specialized databases makes it more readily utilized. For example, time series databases facilitate dashboards and visualizations that illuminate complex data.
Manufacturers also employ separate event-based databases for advanced analytics as well as telemetry databases for data quality and system health monitoring. Learning databases specifically dedicated to AI or data science applications are making automated alerts and predictive quality action management possible.
By filtering data to relevant specialized databases, manufacturers ensure that living, active data is available to support powerful business decisions.
6: Providing rich data from the shop floor to enterprise systems outside the factory
The other pillars make new, rich streaming data available. That information has utility beyond the factory floor. Manufacturers are enhancing the impact of their digitization journeys by expanding their data use cases with an enterprise-wide vision. Enterprise systems that handle emissions reporting, global overall equipment effectiveness (OEE), and material optimization have also found value in event-based data from the shop floor. Strategic decision-makers can turn to these insights for important decisions impacting regulation, economic efficiency, and environmental outcomes.
Keep the goals in mind while focusing on the pillars
Having these six pillars in mind should help avoid surprises. The hope is that knowing what has defined previous successes will guide organizations in planning the next steps in their own development, avoiding sticky roadblocks like incompatible databases and stale data.
Modernizing manufacturing processes is not going to happen overnight. Along that long road, continued momentum and an eye for the global picture of where the organization is headed will make for a more successful evolution.