Electricity and Control October 2024

INDUSTRY 4.0 + IIOT

Data analytics, AI and ML demystified Gerhard Greeff – Divisional Manager PM&C, Iritron and Neels van der Walt – Business Development Manager, Iritron

Neels van der Walt, Iritron.

Gerard Greeff, Iritron.

I n the past few years there has been an increase in the adoption and acceptance of artificial intelligence (AI), if one uses a wide definition: from simple spellcheck, pre dictive text, ‘Because you watched …’ recommendation of shows, to complex machine learning and predictive mod els and simulations. However, in manufacturing, there are still a lot of sceptics who do not understand the basics of AI and do not see how it can be beneficial for their facility. Where companies do understand the basics and benefits, adoption is still slow as the next questions are often (as with anything new), where do we start and, do we have what it takes to begin? So, what are data analytics, artificial intelligence (AI) and machine learning (ML)? Do they offer a magical solution? Does a company need a data scientist to be able to use these technologies? Where do you start with an analytics journey? Here, we try to answer these questions and provide some guidance for the (easier) adoption of AI and ML in the manufacturing and processing industries. The figure below depicts data analytics as a continuum ranging from Foundational analytics to Advanced analytics, and including Descriptive, Diagnostic, Predictive and Pre scriptive analytics. Descriptive analytics Descriptive analytics forms the basis for all other analytics categories. Most companies already have Descriptive an alytics in operation, as this answers the question of ‘what happened’. Software technologies supporting descriptive analytics include plant historians that can record a large number of plant parameters every time the values of the parameters change. Time series data comprises a date and time stamp, and the process value. Time series data is used to find out and report what happened before and after a deviation occurred. Time series data also forms the basis for industrial AI and ML. Other technologies that enable descriptive analytics in clude Manufacturing Operations Management (MOM) sys tems and data tracking, for example: transactional records on material and product movement throughout the manu facturing value chain, the related quality samples and their results, plant throughput compared to the plan and reasons for plant downtime for a specified period, and more. MOM technologies typically include comprehensive reporting to be able to answer the ‘what happened’ question. Diagnostic analytics The next step in the continuum is Diagnostic analytics and this answers the question ‘why did it happen?’. Addition

al context is usually needed to answer this question. Root cause analysis tools bundled together with plant historians allow for the analysis of data by trending various process variables together and viewing the values in relation to one another on the same time axis. This makes it possible for process engineers to investigate process anomalies, ena bling them to understand why things happened in a certain way at a given moment in time. MOM systems are also powerful in providing context as they present time and transactional views related to pro duction, inventory, quality, and maintenance. This context allows the user to gain a deeper understanding and insight into plant or facility performance. As a simple example: looking at production numbers together with shift informa tion, allows the user to compare performance between, for instance, dayshift and nightshift. Other context cate gories may include product, plant, production line, quality, throughput, and seasons. Predictive analytics The next step in the continuum is Predictive analytics. Pre dictive analytics answers the question ‘What might happen in future?’. Predictive analytics is typically based on AI and ML technology and needs the same time series history data as introduced for descriptive analytics. Predictive analytics is ideally suited for processes or assets that are critical for production or quality, or assets that are capital intensive and critical for plant throughput. Commercial off-the-shelf AI and ML solutions do exist, so it is not necessary to un derstand complex mathematical models or to be a data sci-

The data analytics continuum.

OCTOBER 2024 Electricity + Control

7