Everybody is talking about the potential benefits of big data, as if more data is always better. But big data can cause big challenges too. Once you’ve captured data from sensors, you need to store it, plow through it and try to make some logic out of it to be able to use it in your daily business activities. With all the current hype around big data, Industrial IoT and Industry 4.0, your ideas may be big too. You might want to combine data from multiple data sources, various departments and devices to build your cross departmental self-learning predictive model to enhance your decision-making process… So where to start transformation?
Make a plan
Just being able to capture all the data is not in itself a benefit to the business. Before you can reap the benefits, all that big data needs to be turned into actionable insights. This means having a plan. That plan starts with understanding the business itself, the manufacturing process and what you want to achieve and it ends when you enable the big data to contribute to performance optimizations and even new insights. Starting with a plan not only helps you achieve your end goals, it can help you avoid potential challenges to your success.
Know the challenges – and how to avoid them
The biggest challenge most businesses face is time. Current research has shown that the big ideas all too often lead to long analytics modeling projects. In these projects, data scientists rely on various stakeholders to gather input to build their model and deliver the first outputs. That input data might be unavailable, incomplete or ambiguous, leading to disappointing or unreliable results. It may also require reformatting to be ready for inputting into the model. This all takes time, which is why data models are generally used only to address the most critical questions. Only after multiple iterations with data collecting and cleansing projects on the side can the data modelling project deliver actionable information. This can then be provided through dashboards to answer the initial question.
The problem with this approach is the time involved – time from stakeholders, time from data scientists, and the time the organization has to wait for a result. If the project takes too long, the answers it generates may no longer be as relevant to the business… And the organization may have lost its enthusiasm for the results, and made a workaround while they waited.
Why make it so hard? There is a better way
The truth is that there is already an easier way to achieve fast results. The answer already lies within your organization, in the data that has been captured for many years already: your time-series sensor data that has been captured in a historian. This data gives a huge potential to improve production yield, reduce waste, control energy consumption, extend asset availability, predict maintenance and improve regulatory compliance – all it needs is the addition of analytics.
Historians are designed to store data, but not to index and quickly search through it – let alone to visualize it. Luckily, a new generation of technology can solve this need by providing analytics in a self-service form. This means advanced analytics power in an easy-to-use system that is designed for end users with little or no data science knowledge.
Why give this power to end users?
Working with time-series data is best done by the subject matter experts (such as process engineers and control room staff) because they have the knowledge what to look for in case of anomalies in process behavior and finding root causes. They can also identify best performance regimes that can be used as define ideal production and identify conditions for live process monitoring and performance prediction. These subject matter experts are in fact the key to improving the company’s profitability – all they need is the tool.
When you give the power of self-service analytics insights to the people running your plant, you’ll open the door to improvements at all levels of production from day one.
Start small, benefit immediately
The trick for a steady road to success is to start small with analytics, enabling the organization to grow on the path of analytics and increase the scope over time.
This does not mean throwing away your big data – it means focusing on a smaller portion of the data that is the most relevant to your needs to deliver the first results, and prove the value. After a short learning curve, subject matter experts can analyze, monitor and predict more complex use cases for manufacturing process performance. Data from additional sources can also be added, such as energy consumption, weather influences, air pollution measurements, quality measurements, fatigue calculations, etc. to broaden the scope of your use cases.
What are you waiting for?
Take the first step to your organization’s transformation today – it’s easier than you thought.
If you’d like to know more about getting started with analytics, or discuss how we can support your organization’s journey to transformation, please contact us – we’ll be glad to help you.
CEO Bert Baeck cuts through the hype around Industry 4.0 to show the practical considerations of using big data for industrial analytics. Discover the opportunities that disruptive technology offers the process industry, and how to get started. Learn how a self-service approach makes this change easy and achievable.