Smoothing out interoperability issues in smart factories

Ensuring non-stop connectivity throughout converged networks involves dealing with a hodgepodge of protocols and technologies.

Data mining is transforming the face of factory automation. As factory automation is all about faster, smarter and more efficient production, unlocking the vast, untapped potential of data on shop floors has been the driving force behind factory automation’s transformation for some time now. The most striking change on factory floors is the evolution of machine-to-machine (M2M) communication into system-to-system communication. M2M communication still propels automation significantly when it comes to point-to-point communication between hardware, but it takes a back seat to system-to-system communication when data needs to be transferred from sensors and devices to the cloud. Consequently, establishing connections between subsystems in network architectures adds new communications paths and network platforms, bringing new complexities and challenges.

While discussions about the Industrial Internet of Things (IIoT) and its promises of cost-efficient, non-stop operations carry an opportunistic tone among executives, it is system integrators who have to grapple with the reality of figuring out how the different platforms in a network can communicate with each other. What turns a system integrator’s world upside down is the hodgepodge of protocols in the three divergent domains of network architecture: operation technology (OT), information technology (IT), and the IIoT. Each domain comes with its own set of protocols that effectively creates non-interoperable silos, making it impossible for useful data to reach those who need it on an enterprise level to make important decisions and leaving system integrators at their wits’ end. Matters are further complicated by the fact that both the OT and IT departments are unfamiliar with the protocols used in each other’s domains. This trend must be reversed quickly, because as the IIoT makes inroads into automation, OT and IT are converging.

There is good news, though. Solutions are available to bridge these interoperability issues through a variety of protocol conversions. This article sets out to take a closer look at the challenges that system integrators face with OT-to-OT, OT-to-IT and OT-to-IIoT interoperability, as well as the solutions available to ensure non-stop connectivity throughout converged networks.

OT-to-OT interoperability

OT-to-OT communications in factories are not as simple as they used to be. This can mostly be attributed to the IIoT, which has brought a plethora of sensors and machines to the internet on a massive scale. These types of communications are not going to get simpler anytime soon, as the rise in connected IoT devices is expected to continue exponentially. This surge in devices being able to be connected is impacting factory floors in such a big way that M2M communications have evolved into the need for communications between divergent operational subsystems to fulfil data collection and analytics. The snag however is that the heterogeneous systems that fall under OT — such as manufacturing execution systems (MES), supervisory control and data acquisition (SCADA) systems, distributed control systems (DCS), programmable logic controllers (PLCs), and the machines and sensors on the plant floor — all run their own protocols; consequently, the age-old issue of non-interoperability rears its head again and a multitude of protocol conversions are required.

An example of where efficient communications between disparate OT systems on the factory floor benefit operations is having the heater, ventilation and air-conditioning (HVAC) system work in sync with the production system. When the latter’s workload increases, it alerts the former to start up to ensure that production will not be interrupted by overheating or freezing temperatures.

Challenge: An alphabet soup of protocols

The growing complexity of operations processes brings more and more heterogeneous systems into the equation. This means more devices and more protocols. Installation and set-up require more time to plan the architecture and perform device commissioning. For system integrators, it is all about saving time and cost. They don’t want to spend long hours on device commissioning and configuration, or on protocol conversions. However, it is not uncommon for them to spend hours on communication and troubleshooting programming when using communication modules or small PLCs. System integrators therefore want an easy way to simplify protocol conversions so that they can better spend their limited time on their core tasks, such as programming the automation system.

More and more operators are taking advantage of industrial protocol gateways to accomplish the mass configuration of devices and protocol conversions between different devices to keep operations running smoothly. For example, in an electricity room, bridging a large number of Modbus RTU power meters to a Modbus TCP network is usually extremely time-consuming due to the configuration of the slave ID routing table. A convenient solution is to include an automatic device routing function in the gateway that automatically detects the commands from a SCADA system and sets up the slave ID routing table. With only one click, this configuration can be achieved within a minute. Furthermore, a ready-to-use protocol gateway that supports the multiple industrial protocols commonly used in OT (such as Profinet, Profibus, EtherNet/IP and Modbus) simplifies protocol conversions, resulting in significant cost and time savings.

OT-to-IT interoperability

Close cooperation between IT and OT professionals is fundamental to leveraging any smart application’s IIoT platform. Although the approaches of OT and IT to problem-solving differ vastly, they both work towards the same goal: optimised production. To be successful, both domains need access to industrial data. IT departments, which oversee enterprise resource planning (ERP) and sometimes MES, need to review this data to form the bigger picture and then develop solutions for each of the issues that hamper an operation’s reliability. OT professionals are more closely involved with the physical operations on the factory floor and have to figure out how to make all the divergent systems, fitted mostly with proprietary technologies, work together. On the other hand, a positive trend in the era of Industry 4.0 is the OT staff’s increasing recognition of the importance and convenience of IT technology as it helps them achieve their goals.

Challenge 1: The big divide

IT departments face an increasing demand to collect production data from shop floors in order to optimise production. For IT staff, this is not an easy task as they are not familiar with the process of collecting data via industrial protocols. At the same time OT staff members face a similar predicament, in that once they have the need to transfer OT data to the IT layer, IT departments often request interfaces they are not familiar with. This can potentially spark a power struggle between the respective domains of expertise over interfaces and protocols. In the age of Industry 4.0, it is not in any organisation’s best interest to keep the OT and IT domains apart; therefore, eliminating the knowledge gap between them and aligning them more closely deserves operation managers’ undivided attention.

A multiprotocol integrated device will make the lives of system integrators much easier here. For example, a smart I/O device that supports various protocols — such as Modbus TCP and EtherNet/IP for OT engineers, and SNMP and RESTful API for IT engineers — allows communications with different interfaces. This is certainly a step in the right direction to bridge the divide between OT and IT. Such a solution makes it possible for both IT and industrial automation (IA) engineers to conveniently retrieve data from the same I/O device.

Challenge 2: A view to staying ahead

Demonstrating just how much OT and IT are worlds apart from each other is the fact that OT networking devices are always treated as transparent, making it difficult to monitor them — even in the case of emergencies. This adds to the frustration of network operators as troubleshooting becomes almost pointless when they are experiencing downtime. Of course, this situation cannot be accepted, as situational awareness is very important for network operators in order to ensure continuous production and prevent abnormal situations. Ensuring continuous visibility of all network devices and the status of a network in a control room is top priority. However, to capture abnormal events on the shop floor and then interpret the information about these perceived events in real time is quite challenging, due to the complexity of the protocols and networks.

For production lines that employ OT protocols, Ethernet switches that support Profinet, Modbus TCP and EtherNet/IP protocols enable engineers to simultaneously view data and the network status at a central site on a control system, or locally on a HMI. If an industrial protocol fails, the switch reports it and the PLC sends an alarm so that the situation can get fixed immediately.

Leveraging IT’s expertise and knowledge in the domain of Ethernet networking can speed up troubleshooting, reduce system downtime and increase situational awareness.

OT-to-IIoT interoperability

In boardrooms, executives expect data mining and analytics to pay dividends in reduced operational costs, optimised production and predictive maintenance to minimise downtime. As one would expect, this data needs to be collected from field sites, and it has become the job of OT engineers to transfer this data from the devices in the field to the cloud, where it is stored for analytics purposes. This new addition to their job description takes OT engineers somewhat out of their comfort zones, as they would rather focus on programming and work that adds value to their specific field instead of communications tasks.

Challenge: The need for speed

OT engineers’ lack of IT knowledge is definitely their Achilles’ heel. As it is, sending data from an edge device to the cloud can be time-consuming, and OT engineers’ lack of familiarity with IT technology only compounds the process. In the race to IIoT connectivity, the biggest challenge for them is to cut down on the time to set up and program the networking connections between edge devices in the field and the cloud.

To spare engineers a great deal of programming effort and reduce time and costs, an embedded computing platform that supports versatile interfaces, coupled with a software suite that integrates a ready-to-use Modbus engine and cloud connections such as AWS, enables fast integration between devices in the field and the applications required for the IIoT. Furthermore, for those who want to adopt OPC UA in order to unify automation interfaces, software solutions are available that provide an OPC UA server as well as cloud connection capabilities. The beauty of these types of solutions is that they require no additional costs to implement functionality to enable cloud connectivity.

Conclusion

The growing complexity of operations processes and the growing need for IIoT connectivity have become major challenges for smart manufacturing companies. Installation and set-up require more time to plan the architecture and perform device commissioning, and manufacturers and OEMs alike are plagued by a plethora of communication protocols.

Not only is protocol integration and the need for additional data collection a challenge — for both OT and IT personnel — but there is a corresponding need for greater situational awareness around network reliability for operators in both the field and control room in order to ensure continuous production and prevent abnormal situations. However, to capture abnormal events on the shop floor and then interpret the information about these perceived events in real time is quite challenging.

Fortunately, hardware and software solutions are now becoming available that simplify the task of protocol conversion and provide greater visibility into the data-gathering network function so that faults can be rectified quickly.

Top Image: ©stock.adobe.com/chiradech