Digitalization How to build the optimal industrial solution in the era of global IoT Digitalization In today’s heavy industries connectivity is a must to enable services such as remote maintenance on a global scale. However, selecting the optimal technology to establish connections can be challenging in industrial environments. What kinds of factors need to be considered? Share this story: In the era of global Industrial IoT, machines are equipped with plenty of sensors – for good reason. The manufacturers and owners of equipment reap clear business benefits from the massive amounts of data sensors are gathering. This data is invaluable for industrial digitalization. It enables transforming heavy industries into more data-driven businesses. However, having all this sensor data sitting idle at its point of origin is useless. It needs to be transferred out from this location to squeeze out its business value. Connecting operational technology (OT) to IT systems tends to be much more complex than networking offices. “In general, IoT solutions emphasize the fact that there is not only an IT/OT environment. For instance, communications and computing devices must tolerate different weather conditions, corrosive chemicals, and vibration,“ says Risto-Matti Ratilainen, cloud solutions architect, and head of Etteplan’s data and cloud teams. He has a wealth of experience with industrial use cases and their connectivity challenges. “The installation’s location has always an impact upon connectivity. A lot of electromagnetic interference can occur in industrial settings, which can make wireless networks unreliable. Another aspect is having equipment placed in rooms that resemble a reinforced concrete bunker – that is not exceptional at all, and it complicates setting up wireless connections.” In some cases, the key issue is the distant location with poor Internet coverage. “A perfect example of this is our customer Tana that needed to get remote live monitoring of its waste shredders and compactors distributed around the world. The most important thing was to collect data at the edge, bring it to a cloud backend, and provide insight to Tana, its customers and partner network via a web portal.” Navigating between connectivity options Today, there are many different connectivity solutions on the market, and new ones keep popping up. It can be confusing to navigate this sea of multiple technologies. “My experience is that customers are quite aware of all options. Often there is talk about 4G, 5G, LTE-M, NB-IoT, private networks from different providers, and so forth. But customers expect sparring and discussion about experiences to make the right choice,” Ratilainen tells. The first thing is to clarify what is really needed as it affects the choices around connectivity. “There is great variety here. Typically, the primary need is just to get OT/IT integration. Sometimes it is enough to gather time series data with relatively limited network capacity requirements. In another instance the customer wants real-time production data for quality control, monitoring, reconfiguration, and reporting. This can require high-definition video which in turn requires high bandwidth and capacity.” The maturity of any technology is a critical factor in industrial production. It limits the possibilities to use the newest connectivity solutions on the market. Also, data security is heavily emphasized in IoT solutions so that intruders cannot gain access. The price tag of connectivity needs to be considered as well, as the implementation must be cost-efficient. Benefits of edge computing, and digital twins Transferring all raw data that sensors generate rarely makes sense. Otherwise, networks and data storage would become flooded by irrelevant information. Therefore, some level of edge computing is important to compress data before it is sent anywhere. Additionally, IoT endpoints widely support the MQTT protocol, which is well suited for sequential transmission of event-type data traffic. “Especially when using cloud native IoT endpoints one needs to consider how data is compressed and transferred to the cloud. Also, you need to plan data management in the cloud; how long data is stored, how it is accessed, and by whom,” Risto-Matti Ratilainen explains. Regarding the data’s purpose, it is important to think about its value for the whole value network. Often equipment manufacturers want internal telemetry data for error clarification and remote maintenance. Customer access can be provided as an additional service, which enables monetisation of IoT. “Industrial companies need better fleet management at a growing rate. They want maintenance to be easier, cheaper, and more sustainable. Sending people to fix a machine is a cost issue as well as an environmental issue. A hot topic is using sensor data to create digital twins. “There is great potential in digital twins combined with IoT solutions. We already use them in simulations for instance, to do real-time optimization of a wind farm, adjusting blade angles for wind conditions and humidity. Cloud services also provide great 3D graphics so you can visualize the digital twin to customers,” Ratilainen says with enthusiasm. Digital self-service solutions are a great way to leverage value from data in an industrial setting. Read more about how Self-service is taking over B2B!