Skip to main content

AI • IOT • NETWORK EDGE

Case Study: How to Scale Your Industrial IoT Project

Designers of Industrial IoT systems have faced significant interoperability and connectivity challenges since before there was an IoT. These hurdles include integrating new and legacy devices into a unified system, supporting a wide range of connectivity technologies, and dealing with inconsistent Internet access.

Consider the Maersk Group, a Danish logistics company that set out to connect its refrigerated shipping containers. The goal of its Remote Container Management (RCM) solution was to monitor 250,000 containers from warehouse to destination, offering visibility into 140 data points such as geolocation and interior temperature. By connecting this data to a cloud-based dashboard, OEMs and customers would be able to avoid unexpected surprises such as late shipments or spoiled goods.

But there was a problem: The container control units had been sourced from different suppliers. Even controllers from the same supplier often had different communications interfaces or different firmware, making a unified monitoring solution impossible with solutions available at the time.

A new approach was required that could normalize data coming off the varied controllers for ingestion into the RCM back end. The solution also needed to be scalable to support the existing fleet of containers and onboarding of thousands of new units in the future.

Scaling to Success in IIoT: Lessons Learned

The Maersk RCM solution needed an extensible software solution that was device-, data-, and connectivity-agnostic, and also capable of maintaining a consistent data flow across the entire platform architecture. Multiple manufacturers attempted to meet the needs of the RCM project using existing technology components, but largely fell short of the end goal.

Realizing the setbacks of other organizations, two of Maersk’s lead developers on the RCM program learned several key lessons during the course of the project that led to the deployment of a successful solution. These included:

  • Full-scope pilots—Fast, iterative pilots are a requirement of large-scale IIoT projects like the Maersk Line RCM. These allow developers to perform comprehensive testing of the complexity and scalability of a system implementation, as well as the business case.
  • Universal device support—To realize scale at the level of the RCM project, device support should be as universal as possible. Not only will this allow an IIoT solution to proliferate across as many endpoints as possible, capture the most data possible, and derive business value from all assets in a system using a single process.
  • Capture all data—Because data can lead to new potential ideas, applications, and outcomes, as much of it should be captured as possible. This is particularly relevant in the prototyping and implementation phases but should persist throughout the lifecycle of an IIoT system. This requires an extensible architecture that can capture data in many different formats.
  • Frequency versus size—Smaller, more frequent events have the potential to affect an IIoT project more than large, isolated instances. It is therefore essential to capture as much data as possible while building a business case and measure the number of event occurrences against their individual impact over time.
  • User-friendly—The ease and speed with which nontechnical operators can learn and use your technology is critical for mass adoption. This was especially applicable to the RCM solution, as the technology was often deployed in remote locations with on-site personnel who had no experience in networking, communications, etc. Plug-and-play solutions minimize errors and platform deployment time, which accelerates return on investment (ROI).

These considerations helped the Maersk team meet the requirements of the RCM product line, including the ability to integrate thousands of diverse controllers from different suppliers into a unified back-end architecture, support for a variety of connectivity technologies, and a solution for containers with intermittent access to the Internet.

The challenges of the Maersk Line RCM, and the best practices developed to solve them, are common across large-scale IIoT projects in industries such as manufacturing, utilities, oil and gas, and commercial buildings. This realization led to the creation of Omnio and its plug-and-play IIoT onboarding solution.

Connecting Diverse Controllers and Data Models

The Omnio IoT onboarding solution connects with more than 10,000 end device models from manufacturers like ABB, Danfoss, Schneider Electric, and Siemens—including devices like PLCs that were not designed for cloud connectivity. This broad device support enables integration of both legacy industrial equipment and modern IIoT devices, which allows the solution to achieve massive scale.

To extract data from diverse operational technology (OT) devices, the Omnio software is compatible with industrial protocols such as Modbus, Profibus, DeviceNet, Ethernet/IP, and Bacnet. After ingesting inputs from these various protocols, the platform converts the data into IoT-compatible formats such as MQTT, OPC-UA, or HTTPS for transport to local edge gateways or the cloud.

A critical feature of the Omnio gateway software is a normalization layer that standardizes data models and descriptors for upload into a customer’s back-end IIoT monitoring solution (Figure 1). For example, if two separate controllers categorize temperature using Fahrenheit and Celsius units, the Omnio normalization software translates these values into a uniform data set with aligned labels and units. The output from this layer is an enterprise-compatible format such as JSON or OPC-UA.

Figure 1. A data normalization layer in Omnio gateway software standardizes data models, descriptors, and labels, and outputs this information in enterprise-compatible format. (Source: Omnio)

Another piece of core functionality in the Omnio solution is the ability to capture data from devices with sporadic Internet access. Here, the Omnio software supports a feature called pre-commissioning, which allows the platform to start extracting device data in areas with little or no connectivity so that critical information can be retained.

IIoT Simplicity with Intel® Secure Device Onboard (Intel® SDO)

And because personnel on the user side of logistics, manufacturing, or oil and gas may not be technically savvy when it comes to onboarding devices, simplicity is key. To ease the complexity of onboarding devices like container controllers, Omnio relies on technologies like Intel® Secure Device Onboard (Intel® SDO).

Intel SDO is an IoT platform-agnostic feature that leverages the unique Intel® Enhanced Privacy ID (Intel® EPID) and hardware security measures to provision connected systems with the cloud quickly and securely (Figure 2). Available on Intel® Pentium®, Intel® Celeron®, and Intel Atom® processors based on the Apollo Lake microarchitecture, Intel SDO also pairs with a scalable range of industrial-rated processors that offer different levels of performance for IIoT edge computing applications.

Figure 2. Intel® Secure Device Onboard (Intel® SDO) uses unique, hardware-secured device IDs to quickly onboard Apollo Lake-based systems with IoT cloud platforms. (Source: Intel® Corp.)

Continuous Scalability

Unlike consumer forms of IoT, IIoT architectures must support tens or hundreds of thousands of nodes, many of which are legacy, rely on a diverse set of connectivity technologies and data communications structures, and can be distributed across broad geographic areas. IIoT solutions need to onboard these devices easily, integrate their data seamlessly, and provide quick access to value and ROI.

Lessons learned during development of the Maersk Line RCM product enabled 250,000 shipping containers to connect to a unified monitoring platform without having to retrofit each controller with compatible firmware and communications technologies. And because the solution was able to output normalized IP-compatible data packets, information from the container controllers could be transported easily over wide areas and into the cloud. In this way, the RCM product line was able to achieve massive scale regardless of the diversity of end devices, allowing Maersk to grow its fleet of shipping containers to 380,000 over the past nine years.

Today, solutions like the Omnio plug-and-play onboarding solution provide the same level of integration, interoperability, extensibility, and fast time to market, off-the-shelf. The Omnio platform already supports most common device types and protocols, and new ones can be added in a matter of days.

Although no two IIoT projects are identical, careful planning and development to the core best practices outlined above allow organizations across industrial markets to maximize device support, data capture, and ease of use in connected deployments. This is your roadmap to high levels of interoperability and scale.

About the Author

Brandon is a long-time contributor to insight.tech going back to its days as Embedded Innovator, with more than a decade of high-tech journalism and media experience in previous roles as Editor-in-Chief of electronics engineering publication Embedded Computing Design, co-host of the Embedded Insiders podcast, and co-chair of live and virtual events such as Industrial IoT University at Sensors Expo and the IoT Device Security Conference. Brandon currently serves as marketing officer for electronic hardware standards organization, PICMG, where he helps evangelize the use of open standards-based technology. Brandon’s coverage focuses on artificial intelligence and machine learning, the Internet of Things, cybersecurity, embedded processors, edge computing, prototyping kits, and safety-critical systems, but extends to any topic of interest to the electronic design community. Drop him a line at techielew@gmail.com, DM him on Twitter @techielew, or connect with him on LinkedIn.

Profile Photo of Brandon Lewis