Industrial analytics trends

Industrial Analytics Trends 2017

Discover 3 key developments in the industrial analytics market that are driving success, and how to leverage them today.


I’ve just returned from the ARC Advisory Group’s European Industry Forum in Barcelona. Each year this event highlights key industry trends for asset-intensive and manufacturing industries, such as IIoT, Industry 4.0, edge computing and cyber security. This year’s conference had a large focus on industrial analytics, a topic that is transforming more than just the way that we look at process data today.

Looking back over the past few conferences we’ve seen a very positive trend among the early adopters of industrial analytics. In this blog post I’m going to share 3 key factors to their success.

 

Advanced analytics – no longer just hype

In the last couple of years we’ve seen a clear increase in the adoption of data analytics within process manufacturing companies. Just a few years ago, big data analytics for industry was more of a “big dream” than a reality, except for a few early adopters. Even last year when new applications were being more widely shown, the benefits were not yet substantiated and many companies were hesitant to invest.

This year however, organizations are able to prove the value they have gained from using industrial analytics – and adoption is rapidly accelerating.

 

1 – Proving the value of industrial analytics

SitechSitech is a service organization at Chemelot, Geleen, The Netherlands, serving 7 different chemical companies: Arlanxeo, Sabic, OCI Nitrogen, Borealis, DSM, Ancore and Fibrant. Sitech provides these sites with data analytics services via their Asset Health Center.

At the ARC conference, Marc Pijpers of Sitech did a presentation which really highlighted the impact of industrial analytics in practice, and why adoption is expanding so quickly today.

Pijpers pointed out that while big data and advanced analytics are not new for them, despite good results from early analytics projects they needed a faster way to get the value out of the data. Instead of spending 80% of their time in data gathering, cleaning and preparation for analytics models, they wanted to spend 80% of their time getting answers from the data.

That was the point at which Sitech started using a self-service industrial analytics solution. This transformed the way that Sitech was able to analyze process data for their customers at Chemelot. By empowering subject matter experts with immediate answers to daily data questions, Sitech dropped the biggest barrier to analytics adoption: the time and money needed to gain results.

2 – The emergence of self-service industrial analytics

For industrial analytics projects there are now two major approaches.

Data modelling
The traditional and most common approach to industrial analytics involves data scientists building an analytics model. Data scientists must understand the use case and then gather, transform, optimize and load the data in the developed data model, which needs to be validated, optimized and trained to get to business value. The completed data model provides the answers to the initial questions.

Aside from the long time needed to realize results and the high cost, this way of working has another major disadvantage: it leaves companies completely dependent on their data scientists, and results in a solution that the subject matter experts (engineers and operators) may not fully understand.

Do it yourself
The other approach is self-service. This is a new way of working which has only recently become possible. With self-service industrial analytics, there is no need to model data. Companies do not require a data scientist to use the software, and there is no long project timeline or high cost. Instead, the subject matter experts directly query their process data at any time in a self-service application.

By democratizing access to analytics insights, actionable information becomes available at all levels of the plant. This means the ability to achieve incremental improvements at all stages of the production process.

Fabrice Leclercq Rotating Machinery Engineer TOTAL Refining & Chemicals

3 – The rise of reality in analytics projects

The organizations that presented at the ARC Forum this year, such as Sitech, Dow Chemical and Stork, have all achieved success by using industrial analytics – but it took time. All of these speakers had the same advice to new adopters: whether you want to start using model-based analytics or the modern self-service analytics approach, you need to start with small projects and deliver results, rather than tackling the most complex use cases first.

central analytics teamThrough the concept of ‘big data – start small’, your organization will learn how best to apply advanced data analytics to your specific needs. Front-runners in your company (those with the greatest understanding of analytics) can act as a central team and help the rest of the organization to start unlocking value.

The central group gathers valuable expertise from initial projects, which benefits the organization in two ways: the company makes immediate steps into the era of Industry 4.0 and the central group then helps users across the organization in using advanced analytics in their day to day activities.

In this way the entire organization transforms into a data driven organization as quickly as possible, without waiting for a single project to deliver results.

 

Start succeeding with analytics

Analytics pioneers like Sitech, Dow Chemical Company and Stork have proven how to succeed with industrial analytics. With a realistic approach and the speed of self-service solutions, you could start benefiting from these trends immediately.

To get started, contact us – our team will be glad to help you get started.

 

More information?

Sitech webinar on demand

Hear a real customer’s story – in this webinar, Marc Pijpers of Sitech shares his experience of using (predictive) industrial analytics to contextualize asset performance with process data, and illustrates the benefits with 3 actual use cases.