Asset Management Automation HMI-Mobile IIoT SCADA

Leverage Data To Create Your Digital Plant

EP Editorial Staff | June 10, 2021

Combine knowledge asset-operation knowledge with understanding of natural and physical conditions to develop accurate models to which you can apply AI.

These five steps will help you build an operation that leverages digital-twin and AI technologies.

By Matthew Wells, GE Digital

In today’s rapidly changing industrial landscape, companies must embrace digital transformation and create a digital plant to keep up with the pace of change, meet growing operations challenges, and remain competitive. The global pandemic has made this even more of a priority as companies address the need to be more agile in operations to protect their employees and their business.

How do you start the digital-transformation process? This is particularly difficult for industrial organizations due to legacy automation assets, disparate software applications, and the constant need to keep up with ever-changing customer demands. Fortunately, proven solutions and processes provide the foundation for creating digital, optimized plants.

Transformation of any sort requires a clear vision of the future. You also need executive buy-in, financial support, possibly new organizational models, and a cultural shift that drives a new mind-set in your employee base. A holistic approach is vital to seeing this through and ensuring digital transformation is a phased endeavor that delivers sustained results. Transformation doesn’t matter unless you have a culture that’s willing and able to embrace it.

By taking a step-by-step approach to creating a digital operation, you won’t overwhelm your existing infrastructure or your team. Along the way, you can adjust your changes to what works for your organization. It will also give you an opportunity to include management in your journey.

Step By Step

Step one: Start. Sometimes, this is the hardest step to take because you have to identify which piece(s) of your process is/are most critical. In some instances, it’s a small change that has a big impact. Make sure you are not looking for solutions that will only address one niche need. Investment of time and money can provide a quick payback but must provide the building blocks for future optimization.

Organizations need to understand how to configure their operations for transformation, determining what capabilities, roles, leaders, and teams are needed. Also, what are the tools you need to employ to make your transformation successful?

Step two: Connect the data. One thing is certain: Digital transformation starts with capturing operational data, combining it with other meaningful data sources for context, and managing a historical record. It is data, turned into information, that provides the basis for meaningful outcomes.

Companies that effectively leverage data set themselves up for long-term success. A connected and digitally oriented enterprise will allow you to capture more data and make it actionable. Real-time visibility into asset performance across devices can lead to cost savings, improved productivity, and reduced waste.

If you look at this as your base layer of information gathering, you will generally start with an HMI/SCADA solution as the first data source. This is where your operation is controlled, so it is where your operators look for the information needed to increase efficiency. As smart and IoT-enabled technologies become more pervasive in industrial organizations, it’s the data derived from these base systems that will drive the change that companies seek.

Many companies understand the potential and challenges of gathering data, but there is an added layer of complexity and organization that makes that data effective. IT professionals look after an organization’s information processing, hardware and software installs, and monitoring and data collection. OT professionals understand the plant floor, asset functions, and the facility structure. Aggregating data to ensure that IT and OT professionals have access to that data and can act on it, makes this digital transformation effective.

When you start your digital-transformation process, identify assets that are most critical and make sure you are not looking for solutions that will only address one niche need.

Step three: Standardize. By standardizing how you capture, analyze, and report data, you can provide the transparency and understanding that makes that data more actionable. Data is a differentiator that leverages relevant real-time insights to reduce delivery time and increase throughput. It’s about getting visibility into plant operations, boosting productivity, enabling flexibility, accelerating time to market, and meeting customer needs.

Manufacturing execution systems (MES) can enable these benefits through insights derived from data and intelligence already existing within the manufacturing process and supercharging your continuous improvement with on-premises and cloud-based analytics. By aligning resources, energy, and efficiency management in operations, you can reduce inventory; identify gaps that may exist in your process that are hampering growth and efficiency; lower costs, such as energy; and reduce waste. This all happens with data analysis and analytics.

Step four: Leverage AI. Today’s software and storage solutions can organize data, produce reports based on that data, and make information more manageable and readily available. But now what?

Industrial AI (artificial intelligence) consists of a broad set of technologies and applications, including machine learning. AI and machine-learning models provide insights that lead to business-process transformation and continuous improvement. Knowledge of design, operation, and maintenance of industrial assets, systems, and processes, combined with understanding of natural and physical conditions, provide the accurate models to which you can effectively apply AI technologies. By including different sources of feedback data—human, fleet, plant, model, simulation—you can build a model’s competency, mitigate risks, and strengthen predictive capabilities.

It is not enough, however, for an organization to simply make data-driven decisions. It also can be difficult to strategize a roadmap of powerful analytics applications to drive specific business outcomes. Plant management must also empower engineers—the operations domain experts—to help drive analytics that help automate their business.

Step five: Look forward. Integral to how many leading industrial organizations operate today, machine learning leverages data to intelligently automate processes and decisions, learning from past outcomes. We call this a “closed loop” approach in which feedback drives continuous self-learning and automatically makes process adjustments. By standardizing the capture of data, data analysis, and the output that you expect from the system, an industrial AI solution makes information more manageable and more available.

Incorporating machine learning can be a game-changer in creating your digital plant. This is where the process digital twin enters the picture. Process digital twins create models of the “best way” to operate a process in a given environment. In manufacturing, this can be classified as the “golden batch.” By identifying the best way to manufacture a product, plant operators can ensure that they consistently deliver against quality, cost, and volume objectives.

Process digital twins help you efficiently automate repetitive events, free up operators to work on more-critical actions in the process, save costs, identify issues in a timely way, and ultimately improve operations overall. This technology can help industrial companies meet the challenges of fast-changing consumer demand and regulatory requirements. Some of the results companies have realized by employing this forward-thinking technology:

reduced product waste by as much as 75%
reduced quality complaints by as much as 38%
increased throughput by 5% to 20%
increased OEE by 10%.

Of course, new skills are required to deploy this advanced technology. Companies can easily develop their engineers’ capabilities in analytics and machine learning to fully realize the positive impact that the technology can provide. Engineers can use solutions that feature pre-built analytics and combine their own in-house knowledge base about their processes to leverage analytics toolsets. Often engineers have a hunch about how to improve a process, and analytics can provide a means to simulate and confirm before deploying process changes. Over time, your engineers can deploy this capability across multiple plants, which supports enterprise-wide improvements.

As new data-oriented solutions are deployed in the organization, and you streamline your business operations, you will realize your digital plant and empower your company to continuously improve your business results. EP

Matt Wells is Vice President Manufacturing/Digital Plant Software at GE Digital, San Ramon, CA (go.digital.ge.com). Wells has 20 years of experience with industrial automation and operations-management software. He is responsible for leading product management for GE Digital’s Manufacturing and Digital Plant business and has worked with customers in a wide variety of industries including life sciences, food and beverage, automotive, metals, and pulp and paper.

FEATURED VIDEO

CURRENT ISSUE

View Comments

Sign up for insights, trends, & developments in
  • Machinery Solutions
  • Maintenance & Reliability Solutions
  • Energy Efficiency
Return to top