An Introduction to Industrial Connectivity

Industrial connectivity has come a long way since the first time a PLC was controlled by a computer. Well! it was a ‘Hurrah’ moment for industries as it created a whole new horizon for innovative technologies. However, amid the gradual shift towards digitalization, the lack of efficient exchange of data among systems and applications was hindering the communication. When ISA-95 reference model came into light, it compartmentalized the automation architecture into different vertical layers based on the nature of data generated. While this model allowed various industrial manufacturers to innovate technologies keeping the architecture layers in mind, it also helped them understand the communication interdependencies among the systems across the layers. Fast forwarding to today, the coining of the term ‘Industry 4.0’ has emphasized on interlinking various systems (machines, devices, applications, etc.) from plant floor to the enterprise applications of ISA-95 to become a smart factory. This interlinking is possible through efficient connectivity solutions enabling smooth data exchange across the layers. These connectivity solutions are designed keeping the communication needs in mind. While a proximity sensor has a single function, i.e., to detect an object within a certain range, a controller is expected to send sophisticated instructions in different scenarios. Historically, these different communication needs have given rise to the application of various industrial communication protocols.

Factors Influencing the Evolution of Industrial Communication Protocols

As mentioned earlier, the evolution of industry protocols goes back to various scenarios that led various industrial associations and independent OEMs to develop various protocols. Some of the factors that influenced the emergence of various modern protocols are:

• • Interoperability: With generations of electronics and technologies evolving over the decades, the industries started facing difficulties in establishing compatibility among the heterogeneous devices at various layers, especially at the OT level. The devices developed by different manufacturers supported either vendor-specific proprietary protocols or Commercial-Off-the-Shelf (COTS) protocols. Due to this, the need for establishing interoperability among the devices became one of the primary concerns for smooth connectivity from plant floor to the enterprise layers and beyond. This generated the need for common platforms like OPC UA that allows all the devices to communicate in a common language unlocking the potential of IIoT.
  • Real-time/Determinism: When it comes to communication, industries need connectivity solutions that enable fast responsiveness, ensure real-time delivery of time-sensitive messages, and reduce jitter. The OEMs and various protocol consortiums are constantly working to innovate solutions for aforementioned criteria and more. In fact, communication standards and protocols like TSN (Time Sensitive Network) and Profinet IRT are already making significant progress.
  • Operating Environment: One of the most discussed aspects in industries is the safe operating conditions on the plant floor. While some nodes may exhibit a certain amount of heat, vibration, or noise, others may operate in a hazardous environment. Therefore, having stable connectivity channels for such scenarios has always been a challenge. For example, PROFIBUS DP is suitable for manufacturing, whereas PROFIBUS PA has dominated the process industries. In fact, the recent developments in Added Physical Layer on Ethernet (Ethernet-APL) promise to deliver better communication speed along with intrinsic safety benefits to process industries.
  • Mobility: As the plant operations get more complex, newer inventions replace the legacy systems. For example, the use of Automated Guided Vehicles has minimized the number of workers needed to transport materials within the plant. However, the use of wired connectivity does not fulfil the communication need here as the plant asset is mobile. The evolution of wireless protocols has helped overcome this issue. 5G technology will not only allow plant devices to communicate faster than a human possibly can, but will also ensure the delivery of time-sensitive message by slicing the bandwidth.
  • Scalability: As and when industries scale-up, new nodes/devices/machines are added in the network. However, expanding the network always puts a challenge in terms of additional configurations, implementation overheads, implications on existing network architecture, etc. This is the reason why self-healing wireless networks like ZigBee are designed.
  • Power Consumption: With multiple machines deployed on a plant floor, connecting them using specific protocols consumes a lot of power. As a matter of fact, the devices that are battery-powered or electric-powered, a single fault in the power source can seriously damage the entire connectivity. This can be especially a crucial aspect when an end-node is installed at a remote location. Therefore, the invention of low-power wireless networks like Bluetooth low energy, Wi-Fi, etc.

While the conventional purpose of the communication protocols was to provide seamless connectivity among the devices, digital disruption in industries is demanding more than that. The panorama of modern industries needs smooth convergence of OT and IT, which were two different worlds altogether. Along with intelligent devices, industrial protocols are bridging this gap.

How Communication Protocols Converge OT and IT?

Industrial automation pyramid with all 5 layers is a way to look at the communication happening within the system. However, it is not necessary to have all these layers as part of all the industrial network architectures. Since the advent of edge computing, industries are actively deploying it to bypass all the middle layers between control layer and the cloud. This means that the automation pyramid is reduced in size, or in other words, it is flattening, i.e., from 5 layers to just 2 or 3 layers. However, if you look closely, the role of seamless communication is quite important at the moment. While field devices release data at a higher frequency in smaller sizes, client applications on cloud require larger messages in low frequency. Therefore, the connectivity solutions must fulfill the necessary demands of the end industries. In the light of convergence, the role of communication protocols can be discussed at two levels:

Field to Edge

Field devices like sensors and actuators need communication protocols that allow them to communicate in robust way. Some of the communication protocols that are widely used on the field level to connect various machines and devices are IO-Link and the fieldbus protocols like Modbus, HART, Profibus, FF, and Control Area Network (CAN). In fact, Industrial Ethernet protocols like Profinet, EtherCAT, Ethernet/IP, etc., offer great potential to the complex and field devices network. The data transferred to the control layer gets processed and sent to the above layers or specific instructions are sent to the field devices. Therefore, the communication protocols should enable scalability. Some of the communication protocols that provide a scalable connectivity from the PLCs all the way down to I/O and Sensors are EtherCAT, Profinet RT, Powerlink, IO-Link, Modbus, Ethernet/IP, S7, MELSEC, etc.

Edge to Cloud

Conventionally, the data coming from the field and control layers get converted into enterprise-compatible format. However, communication protocols like SigFox, OPC UA, TSN, MQTT, AMQP, etc., are enabling communication right from the sensor to the cloud. The field level specifications of OPC UA, called OPC FLC is under development that will redefine the communication across all the layers of automation pyramid.

Endnote

While connectivity is making a major progress in the industrial front, the OEMs are constantly on their toes to cater to the communication needs of the end industries. With varied demands of diverse industries, there is surely not one communication protocol that can fulfill them all. However, with continuous research and global consortiums coming forward, we can surely expect an influx of innovative technologies paving the way for seamless and improved communication. Utthunga is one of renowned names in industrial protocols that enables the various industry OEMs to engineer cutting-edge connectivity solutions. We are experts in providing device-level and software-level connectivity services along with verifying, verifying, and certifying the solutions at each step. Therefore, let us collaborate to help you fulfil your connectivity needs. Check out our Industrial Connectivity Services to know more.

Building blocks of this tool are :

1. Widgets
2. Dashboards
3. Templates

Widgets: This is the basic component of the dashboard tool. It has configurable elements like Title, Type of Chart, and other options. These widgets can be resized to fit a specific layout and moved around the dashboard to customize the display.

Dashboards: It is a combination of one or more widgets that provide statistics of configured motors, sensors, or other components in the plant. The dashboard can be customized to suit specific requirements in terms of features, functionalities, or visualization layout.

Templates: These are the industry-standard formats used for aggregation and display of data for individual field devices or the entire plant. The Administrator of the dashboard builder can create such templates based on preferences and requirements at various levels such as an operator, plant supervisor, or the plant head.

The primary javascript plugins used for this dashboard builder tool are:

React Grid Layout (RGL)
The RGL system is used for rendering multiple widgets in the dashboard. This helps layout mapping based on breakpoints. It provides intuitive and easy-to-use layout features for dragging and resizing the widgets that enhance the efficiency and responsiveness of the entire application.

Uplot
We experimented with different types of data visualization charting tools such as Chartjs, Victoryjs, and Uplot for rendering a large number of data points. Finally, based on the best time-series data rendering performance, we selected Uplot. With more than 1 Million points to be rendered, Uplot performed intended functions very efficiently.

Plotly
Other than time-series data, we also used 3D mesh plots and indicators for building effective data statistics features in the tool. Among multiple open source libraries, we used the Plotly library. This provided an excellent set of plots that render simple, yet insightful information for detecting anomalies.

React Table
For certain widgets, we wanted more than just regular table features like sorting (client/server-side), footers, and pagination. Among various options, we chose React Table plugin for its versatile features. We have used the standard list as well as the embedded table in the widget that gives the complete solution.

React Calendar / Date Range
The Date-range option is a very common, and also an important feature for any dashboard. For our client, we introduce predefined options for the shortcuts like last-1 Hr, last-5 Hrs, last-12 Hrs, last-1 Day, last-7 Days, and so on for capturing real-time data. Also, the custom date-range option feature for viewing historical data is a crucial dashboard feature. We found the React Date Range plugin an excellent fit for the use cases.

React Filters / Select
Searchable filters are the obvious choice for long data tables or reports. In our case, however, we needed a dynamic searchable component with intuitive selecting features. The React- select plugin provided us with the exact functionality that suited our requirements. On focus, it displays default drop-down options. Also, powerful features like search functionality with async data, and the color options matched nicely with the default bootstrap theme.

 Foot Note:

The Dashboard Builder Tool was developed within a short span of 2 months. The application is live in the client’s production environment, delivering delightful performance.

Utthunga cherishes innovating value-added solutions for its customers in various fields of industrial automation. For your queries and requirements write to us at [email protected].

Whether it’s called Industrial Internet of Things (IIoT), Industry 4.0, or Digitalization, companies have started leveraging technologies to completely reimagine their business model. This transformation will be widespread, and everything from Information technology (IT) to engineering technology (ET) to operational technology (OT) services will be impacted.

The most notable benefit of IIoT is that the focus of business owners has started shifting to increased efficiency and lower costs by using edge devices and Big Data analytics. However, massive data transmissions from and to the IIoT devices in remote areas, on diverse network connections on time are some of the challenges to overcome. Another challenge is the need for actionable data to make sense of all the data collected.

The addition of Edge Computing to the Industrial IoT gateway hardware allows the Industrial IoT gateway to perform local storage and data processing. It results in higher reliability, lower latency, and more security.

There are five widely recognized criteria for choosing the Industrial IoT gateway:

• ​Open standards
• Flexible architectures
• Cloud technology
• Edge Computing
• Cheaper and more flexible hardware​

Why should I use edge devices in Industrial IoT?

The edge device plays a vital role in collecting data and creating a digital twin of your cloud applications. Simply put, the edge devices connect the data from your network to the cloud solution.In a manufacturing facility, edge devices have the following benefits:

• Facilitate condition-based monitoring to keep an eye on the condition of shop-floor machines
• Monitoring and analyzing data to detect anomalies sooner so that you can improve machine uptime, lower spare parts inventory
• Adds efficiency and self-monitoring analysis

IIoT systems with edge devices give you the benefits of real-time local analysis and robust cloud-based storage. Compared to cloud computing, which requires network connectivity and reliance upon 3rd party security, Edge computing, on the other hand, offers low-latency, dependable edge computing power that can be deployed in manufacturing facilities with limited network connections and also in places where security is a premium factor.

IoT device management in IoT platforms – Features, deployment, and capabilities

IoT device management platforms provide device lifecycle management functionality connected to the deployment and management of IoT devices. You can choose a device management platform vendor like ARM, Amazon, Bosch or decide to have IoT platform vendors like Utthunga with device management within their features.

The onboarding process, provisioning, authentication, deployment, and encryption of device management consists of secure and rapid deployment keeping in mind security and trust.

Three key capabilities of IoT device management platforms

1. Deployment and monitoring
2. Maintenance and management
3. Software and firmware updates

Edge device management and its challenges:

As per a report by Statista, 75 billion edge devices are predicted to be used in operation across the industries like manufacturing, healthcare, finance, and more to produce, capture or analyze data. In some cases, these edge devices are expected to capture, analyze, store and transfer data to centralized data centers or the cloud.

The significant functionality of these devices goes beyond managing the captured data as per their software specifications. They are expected to be a part of a larger, interconnected ecosystem to provide users with visibility into entire processes.

Take the case of home automation.There are multiple edge devices for monitoring and controlling lighting, window blinds, temperature, etc., qualified for edge computing in home automation systems. These devices handle specific processing tasks as individuals and act as an interconnected system as more computing power is required for managing edge devices.Limited scalable computing power is only one of the significant challenges of managing edge devices.

Apart from this, enterprises or domestic users need to ensure that these edge devices are secure, even for devices that connect to external data centers. For smaller edge devices, users face are unable to access the right data and get an overview of the work they do. Some form of data visibility is essential when their performance is going to affect an entire system.

Meaning, edge device management requires an all–in-one solution that offers scalability, visibility, and flexibility. A 360-degree view into edge device performance is essential to stay updated with real-time insight.

Industrial IoT platforms and edge device management

Industrial IoT platforms integrate artificial intelligence and machine learning algorithms to take edge device management to the next level. With these algorithms, shop floor requirements can be achieved using extensive data analysis and data-driven insights. IoT platforms also facilitate communication workflows for edge device management which provide deeper insight into connected devices. This communication workflow offers alerts and notifications to help organization’ decision-makers optimize edge computing processes. The data visualization by IoT platforms also makes the comparison of shop floor data against benchmark data possible.

This communication workflow process starts from collecting data from different facilities to create an optimal benchmark of data. Data collected from shop floors can then compare their productivity levels to the benchmarked data, and the insight from this comparison is helpful in optimizing asset utilization. These insights can also be used to highlight other issues and failures within production cycles, right from poor material handling systems or even the idle time of the machines leading to underutilization of the shop floor operators.

Industrial IoT platforms often work as Software as a Service platforms (SaaS) platforms and provide the APIs, algorithms, and repositories for enterprises to help them build edge applications. These applications can also be deployed within edge devices or edge networks, depending on what you want to build.

IoT platforms use the scalability and flexibility of cloud and edge computing to help enterprises scale up their edge device management needs only when more devices or data are produced. Using the IoT platform for edge device management can reduce the cost and efforts as well.

Conclusion

Edge device deployment and management are an essential part when you leverage edge computing within shop floors. The choice of an edge device management platform depends upon your industry’s specific needs, whether you want to build industry-specific applications, cybersecurity, scalability, and subscription cost, or something else.

Edge devices do several jobs, contingent upon what sort of device they are and the use case. Some essential functions of edge devices are the transmission, directing, preparing, observing, sifting, interpretation, and storage of information passing between networks. In IIoT networks, edge devices are network edges infrastructure products like gateways, routers, switches, WAPs, and higher-level end devices like controllers, HMIs, drives, etc.

The software running on the edge device, i.e., the “edge stack” that makes all of the above possible, is made up of proxies connected via network protocols. Simply put, “edge device stack” is the set of brokers that lies between your application and your end-users.

How do Edge Devices Work?

An edge device connects your OT network to IT network and enables you to collect data from all the field devices, process them, and send them to the cloud for further analysis. Edge computing delivers a range of benefits and makes it attractive to industrial/manufacturing organizations.

Industrial edge computing brings low latency computing to manufacturing facilities and is beneficial for organizations using edge computing in IIoT devices. Edge devices’ working principle is that it serves as network entry or exit. It connects two networks by translating one protocol into another and creates a secure connection with the cloud.

Where are the Edge Device Stack Required?

One can deploy edge device stack in instances where they need to use edge devices, and edge devices are used for the following purposes:

• Where there is poor connectivity of IIoT devices
• You have raw data at the edge that need pre-processing to reduce computation
• You are running applications dependent on machine learning and a large amounts of data are required
• You have to keep the data within the factory premises for maintaining security and privacy

By integrating edge computing in the devices and the processes that drive automation, IIoT can enable a paradigm shift in automating industrial processes. The proliferation of edge devices increases the overall surface for networks, cloud data load, data security, and connection optimization, and much more.

Edge devices vary in terms of physical form and capability. Since edge devices serve different purposes, they come in various shape sizes, functionalities and go beyond RFID tags, temperature detectors, and vibration sensors. Intelligent edge devices in manufacturing facilities can include vision-guided robots or industrial PCs.

So, To Buy or To Build?

Whether you should build or buy an edge device stack depends upon your organization’s requirements and the in-house resources you have. Let’s look at the both the options one-by-one!

Building the Edge Device Stack

If you can achieve the customization to suit your unique requirements, going for the build can be the best bet. If you decide to build, keep the following things in mind:

• You have to invest the time, product, and engineering effort building the device stack, as it is complex and diversified in nature
• The infrastructure relationships you have to maintain;
• The resources you will need to hire to build or maintain the device stack to operate the networking, compute, and orchestration technologies
• If you are a build strategy in an area that is most often outside the core business.

If you are clear with the things mentioned above and ready to build your edge device stack, take inspiration from some pioneers of edge computing like Netflix, eBay, and more. All these brands have invested heavily in building edge engineering teams and technology to meet their data, latency, and availability needs from the device edge stack. Keep in mind that edge devices on their own are prone to cyber-attacks and hacks. Therefore, pay special attention to this aspect while designing security architectures.

Pros of Building Edge Device Stack

Building an edge device stack from scratch has its own set of benefits. Some of the pros of choosing build option are:

• Huge predictability
• Maximum control and flexibility
• Efficiency in operation
• Control of the entire process and speed

Every coin has two sides, and building your edge device stack is not different.

The time, effort, and investment you have to do for making your edge infrastructure, including orchestrating resources alongside managing scaling and monitoring, is practically too much to handle. Or put, it might not be feasible for many SMBs as they do not have experienced professionals or that much budget.

Buying the Edge Device Stack

If you are not very convinced with building the stack from scratch, reach out to consultants and partners of the to buy the edge device stack. They will help you implement the edge device stack in the very least effort and time. However, if you don’t choose a competent one, it might cost you even more money to fix the issues they bring. Therefore, do your research on them before you sign a contract.

For example, Azure stack edge is a purpose-built hardware-as-a-service that provides you with quick, actionable insights at the edge where data is created. They offer an easy ordering process and fulfillment. You can order from the Azure portal in a hardware-as-a-service model and pay monthly based upon your subscription to Azure.

Pros of Buying Edge Device Stack

• You don’t have to take the headache of managing the upgrades and adding new features to the stack to adapt to the changing market scenarios
• The support experts at the vendor’s side take the responsibility of configuring it to suit your business needs
• No need to hire specialized resources deploying the edge stack to your network Secure and reliable in terms of proven functionalities, version management, security provisions, etc.
• The total cost of ownership is lesser

The answer to whether you should build or buy an edge device stack depends upon the type and cost of IIoT devices you want to use. If your requirement for commercial software is 60% or more, go for the buy option.

Conclusion

Whether you decide to build or buy the edge device stack, an important aspect to keep in mind is if you can strongly vote for either option. While both have their demerits, they have equally bright benefits too. Fulfilment of your business needs, reliability, feasibility, ROI, etc., are some of the most important factors to scrutinize before going for any option.

Utthunga offers advanced and comprehensive solutions that align perfectly with your organizational goals.