The Quickest Route to Intelligent Manufacturing
It is strongly believed that Industry 4.0 is the key to higher efficiency as it will strengthen monitoring and ensure better management of resources, among other things. Forbes lists the ways through which an industrial organization can become an intelligent manufacturer.
First focus on solving problems faster
When industrial organizations begin to adopt Industry 4.0 technologies to become intelligent manufacturers, it seems that too many focus on the hype and what’s possible in the future. They start thinking about how to connect and integrate every “thing” in the plant and try to implement projects they think will disrupt their industry. Others first consider which technology they should deploy. Ultimately, too many move forward without determining the goals they want to achieve or problems they’re trying to solve.
While it’s essential to have in mind a vision of your future factory, that’s the goal, not the starting point. This new manufacturing era will no doubt give rise to entirely new operations and business models, just as with previous industrial revolutions. It will involve the connection of every machine and operations or business application throughout the business, from design to delivery, and include intuitive human-machine interfaces. Ultimately, the Industry 4.0 organization will seamlessly integrate cyber and physical systems. The question manufacturers have to answer is: “What’s the fastest way to become an intelligent manufacturer?”
Many companies that are leading the race to digitize their businesses say the answer is to begin with targeted, high-impact projects focused on data. More fundamentally, they advise that manufacturers start by identifying how the new technologies will help solve problems faster. As Spotify cofounder Martin Lorentzon once said, “The value of a company is the sum of the problems you solve together.” In manufacturing, if you can solve more problems faster than your competitors, you will reap higher margins, while charging lower prices and accelerating delivery – all while becoming more efficient. Increasing the speed of problem-solving is crucial for manufacturers to get to the next level.
Becoming an intelligent manufacturer is about having people make better decisions and take actions faster. To begin the journey, focus on solving existing problems. Soon after that, you can start addressing issues you wouldn’t be able to without advanced analytics – but that’s the second stage of your journey and the topic of the next post. In practice, both require creating a data architecture that ensures that the right people get the right data and information at the right time.
Data Analytics
The good news is that your facility has all the data you need to get started solving problems faster. It’s locked in existing systems, which include sensing and manipulating: PLCs and sensors; process monitoring: HMI and SCADA; production: MES and MOM; and business planning: ERP and others. The first step to becoming an intelligent manufacturer is to connect these systems, then model, structure, store, and analyze the data from them. By delivering this high-level information in intuitive, easy-to-use dashboards, shift supervisors and operations vice presidents will see what the problems are and what’s causing them. With this information, they can quickly determine opportunities for improvement.
Consider also that people don’t have time to analyze vast quantities of data. Therefore, the more you can automate and deliver data-driven reports, the faster people are going to be able to understand the problem that needs to be solved. With a data architecture that automatically analyzes the quality or capability of the products as you’re making them, filters for data showing what’s bad, and delivers the information via reports and alerts, the engineers will always know what problem they should be solving. They’ll also have the data behind the problem so that they can solve it more quickly.
Solving Existing Problems Faster
Two excellent first use cases illustrate how data analytics can help solve existing problems faster. With the operations and business systems connected and the data automatically aggregated and analyzed, manufacturers can create early warning systems for issues that consistently arise in production, including:
Anomaly detection: By applying an algorithm that reviews production data, the system will identify data points that vary from standard, which indicates a problem. Because the issue is identified and operators alerted in near real-time, the problem can be solved before it’s built into the product.
Predictive quality: An algorithm that analyzes quality data can predict the quality of a part being made about five minutes before any quality defect would be realized. With this information, operators can understand the problem and make the necessary changes to avoid the predicted quality problem from happening.
Vital First Steps
Granted, these use cases aren’t as breathtaking as the prospect of a lights-out factory. However, they can be implemented quickly and deliver a return on investment in one to three months. More importantly, these use cases represent the vital first steps in the journey toward intelligent manufacturing. Solving the low-hanging fruit of reducing variability, addressing quality and machine problems faster, and having a system identify and alert operators when issues happen in real-time lays the groundwork for more advanced analytics – and real digital transformation.
Beginning your journey toward becoming an intelligent manufacturer must start with your data architecture – how you structure and store data. When you start by solving existing problems, you can also focus on creating a robust system where data becomes an appreciating asset. In the past, the more data you added, the slower the data processing became. By setting up the architecture correctly, data becomes an appreciating asset – where the more you have, the more value you’re going to get, because the marginal cost of storage is so low, and the performance doesn’t degrade. This modern architecture allows you to build the massive data sets that’s necessary for building predictive models – and to solve problems that you otherwise would not be able to.
This article was written by Willem Sundblad from Forbes and was legally licensed through the NewsCred publisher network. Please direct all licensing questions to legal@newscred.com.
