SafetyChain

A Guide to the Data Renaissance in Manufacturing

Blog Author Favicon
Roger Woehl & Cameron Bergen
Contributing Writer

While many process manufacturers understand the potential benefits of digital transformation, often, their vision is clouded by past failed projects. Just 30% of digital transformation initiatives are successful and because they require both time and financial investment. 

Fortunately, we can learn from past mistakes. With a little extra planning and preparation, you can pursue a foolproof digital transformation that supports positive change throughout your facility and sets you up for continuous improvement. Avoid the frustration of attempting a digital transformation and failing by adopting a data-first approach. 

What Is Digitization in Manufacturing?

Digitization is a component of Industry 4.0 or the Data Renaissance. The four industrial revolutions include:

  • Industry 1.0, which included mechanization and the introduction of steam and water power

  • Industry 2.0, which encompassed mass production assembly lines using electrical power

  • Industry 3.0, which featured automated production, computers, IT systems, and robotics

  • Industry 4.0, which will be characterized by “The Smart Factory,” including autonomous systems, IoT, and machine learning

At the most basic level, digital transformation is the process of moving business from paper to a digital format. This will then allow your plant to adopt a data-first methodology, in which you prioritize data to drive business efficiency and profitability.

A data-first methodology is a data-driven approach to problem-solving. Its goal is to replace speculation, hunches, and opinions with facts obtained through reliable data. The data-driven process sets the goals and priorities for a successful digital transformation, ensuring a more efficient and profitable business.

What Are the Pitfalls of Digital Transformation in Manufacturing?

To be successful in any digital transformation, you must first be able to answer the following questions:

  • How do we collect data? From where? Who will be responsible for overseeing its collection?

  • Where will we store it?

  • How will stakeholders access it?

  • How will we share and use it?

  • How will we act on the data we uncover?

Clearly, there are some mechanics involved in digital transformation projects, but the fundamental question that must always be answered is the last—how will you make the data actionable? Often, this is where the gap in implementations occurs. Here’s a closer look at some of the most common pitfalls:

  1. Data without a clear purpose does not create business value. In other words, simply collecting data for the sake of it will not yield any meaningful results.

  2. Digital transformation can add complexity that the desired outcome must justify. Companies will find that without an important goal behind digitization, managing new technology and the data it produces may seem overwhelming.

  3. Data automation alone may not tell the full story if it isn’t combined with human data. There may be critical pieces of your manufacturing story that only your operators on the plant floor know.

Collecting the data is just the start, not the end. Analytics and future use are where value exists.

The Data First Approach

To take a data-first approach in your facility, ask these questions:

  • What problem am I trying to solve ultimately?

  • What will data help me to understand that problem?

  • What are the best sources of information?

Keep in mind that something like “improving efficiency” isn’t a problem to be solved; instead, it’s a potential outcome that can come from a data-first approach. If you’re experiencing inefficiencies in your lines, you may need to dig a little deeper to find the root cause, which brings us to our next point.

The First Problem to Solve

Often, the first problem to solve before you can pursue digital transformation is discovering what the real problem is. Using data-driven tools that identify the problem is an excellent place to start. Methodologies like OEE (Overall Equipment Effectiveness) and SPC (Statistical Process Control) can also be used to help uncover the root problems you didn’t know you had. 

How to Think Data First

As you begin to adopt a data-first mindset, here are a few other questions you can use to get a better understanding of the critical issues you’re looking to solve:

  • How does data compare between lines and locations?

  • How does data compare between operators or shifts?

  • How does the data point correlate to the outcomes?

Keep in mind that discrepancies present the perfect opportunity to ask “why.” For instance, why might one shift produce consistently better results than another?

By taking a data-first approach, you’ll be able to unlock your ROI and identify:

  • Gaps in data collection or management

  • Benchmarking

  • Root issues to performance variability

  • A prioritized list of things that can be improved

Of course, machines are only one part of the complete data picture. Data collection by people who are on the ground will still bring insight. To achieve analytics that results in high volume, high velocity, and big data you can use to improve your facility, both human data and machine data are essential.

For instance, you may use data from sensors on your lines to help detect issues and ensure operations are running smoothly. Your sensor data could tell you that the temperature in your ovens is getting too high. That way, you can step in promptly to prevent more significant issues from happening. Although the sensor alerted you to the problem, you might also have to combine this information with human data. Your operators may be able to tell you how often the thermometer has been calibrated, which could be a key piece in solving the underlying problem and performing a thorough root cause analysis.

Why Take a Data First Approach in Manufacturing?

Data is a hot commodity, so much so that some sources are even calling it “the new oil.” To truly envision what data can do for you, think like a CFO. Identify the actual cost of quality, and think of data as an investment that will pay off at a later date by helping you improve your quality outcomes. Data is an asset. While it may not be as evident as a machine or other physical investment, envisioning its function to assist other assets in working faster and better will result in the most significant benefits from your transformation.

Playing the Long Game

The value of data will only increase over time. Right now, leveraging it as a rich source of competitiveness will put you miles ahead in the game. Very shortly, it will be not just valuable to have but essential to business success.

Moreover, data will unlock the door to AI, IoT, and machine learning opportunities further down the road. Often, eager companies jump into these advanced data projects too early, and their implementations ultimately fail. Before you can begin using tools like AI, you must first have a mature data collection process.

The Data-First Culture

The data-first methodology is as much of a cultural transformation as it is a process. To support this cultural shift, leaders will need to model data-first thinking. Data-first initiatives are most successful when driven from the top down. Leadership can navigate this by offering concrete numbers that back up their objectives. At the same time, they must also listen to feedback from management, operators, and others at various levels in the plant, allowing them access to critical insights that may be useful in driving meaningful change.

In Conclusion: Share the Data-First Vision

As a leader, you communicate to your teams that data will be used for driving decisions. The goal is to collect the right data to answer questions and get agreement and buy-in early on to avoid challenges to the validity of the results later down the road. As with any kind of change, there may be some friction initially, but as teams begin to see the efficiencies data can help create, they’ll likely support the transformation.

Additionally, keep in mind that your data should surprise you. Be ready for it to challenge your conventional wisdom and encourage you to break away from the line of thinking, “We’ve always done it this way.” Data could just prove something you have always suspected but could never be sure about.

Finally, be a data consumer, and know that data is not just for the analysis. Real insights rarely fit neatly on one page, so be ready to interact with data firsthand. Ask questions about it, and export the numbers to an interactive application. And, keep in mind that data is for everyone, from the folks on the line to the C-suite. There’s a data element to every role. Making data accessible and actionable for everyone is where actual change can take place.