Identifying the traps that can impede OT data acquisition

29 August 2021

Oliver Wang highlights the traps that can impede OT data acquisition, slowing down a digital transformation.

Industrial digital transformation seeks to break down the silos between an enterprise’s information technology (IT) and operational technology (OT), translating the physical behaviour of OT devices into digital data, and distilling insights with the help of IT’s analysis.

Through OT/IT collaboration, these actionable insights can optimise the overall physical operational system. For example, stance, data from MESs in factories can be integrated with the data from a customer relationship management (CRM) system to shorten time-to-delivery, expand production capacity, and reduce costs, among others.

However, according to the latest Industrial 4.0 Maturity Index, 96% of studied enterprises are still starting out on their digital transformation journey, while just 4% of the enterprises have attained the ‘visibility’ or ‘analysis’ maturity stage. 

Three traps that impede OT data acquisition include:

Invisible environmental traps: Imagine your OT data comes from a drilling well in the middle of a desert where temperatures range from 40 to 50°C, an oil pipeline system that stretches for hundreds of kilometres in freezing cold areas, a transportation system of a fast-moving train that deals with high levels of vibrations, a chemical fuel tank, or a switchgear system inside an unmanned high-voltage substation. A variety of environmental interferences, such as extreme temperatures, vibrations, airborne chemicals, and electromagnetic radiation, can easily cause the malfunction of OT data-acquisition electronics, which results in data transmission instability from time to time, or worse, data inaccuracy, which leads to errors in analyses later on. For example, the large automated warehouse systems in smart factories generate strong electromagnetic interference at the moment of startup, causing anomalies in the network equipment nearby. 

Network disruption, even if just for a second, can wreck the accuracy in the calculation of incoming inventory as well as the production process of the entire product batch.

Unexpected design traps: All OT equipment – from sensors and controllers to control systems – have one thing in common: it is intended to enable highly specialised industrial applications. By design, industrial equipment is for a special purpose. Controllers and sensors used in a drilling well are not the same as the ones supporting power monitoring devices, for example. But if you want to understand the correlation between control levels of a drilling well and power consumption, OT data must be collected from a variety of specialised equipment. Only now, most people realise that every device uses a specific communication protocol that only it can read and understand. Therefore, to glean OT data from more sources, it is necessary to first acquire the ability to ‘talk to’ different devices; otherwise, it makes it much harder to analyse diverse OT data, and it entails direct cost increases.

Data identification traps: Data generated from OT equipment or systems is mostly raw data, meaning it does not come with context. For example, PLCs collect temperature data from sensors deployed in a variety of locations to support monitoring. When the temperature goes above 45°C, fans will be switched on to bring the temperature down. However, for OT data analysts, the raw OT data captured directly from the PLCs lacks context, as they cannot tell what devices were sourced, the data acquisition time, and data owners, etc. This raw data is nothing but a meaningless value in their eyes. Therefore, pre-processing raw data and giving it a context is a key aspect in OT data acquisition. To achieve this, OT equipment vendors must incorporate IT capabilities in their development focus as IT user habits are very much involved in the pre-processing of raw data. Furthermore, if there is too much data, data analysts will be drowned in the sheer workload of converting the collected data into a uniform format for a database, which, fortunately, can now be facilitated by data transformation technology.

OT data plays a deciding role in whether an industrial digital transformation endeavour takes off or crashes before getting off the ground. Before starting a project, it is wise to take stock of the different ways of obtaining OT data and the types of OT data available, as well as plan for converting the data into the format and contextures required by the IT database. Avoid these three kinds of traps and align with your company’s needs to strengthen the OT data acquisition ability ahead of time. In doing so, IT/OT convergence will effectively have to be sped up to allow you to take the first step towards industrial digital transformation, firmly and steadily.

Oliver Wang is product marketing manager, Edge Connectivity and Computing at Moxa. 

Print this page | E-mail this page