Sparkplug: The Open Specification Critical to Achieving ROI in the Industrial Internet of Things

Introduction

As Industrial Internet of Things (IIoT) deployments proliferate and mature, they are a source of service innovation, increased productivity, streamlined operations, and, perhaps most importantly, reduced costs for industrial and manufacturing organizations worldwide. Some companies are expanding their use cases, while others are just getting started. Grandview Research estimates the global market for IIoT stands at 263 billion and will grow to 1.11 trillion U.S. dollars by 2028 (Figure 1).

COVID-19 and the ensuing chip shortages severely hampered the automotive manufacturing industry over the past couple of years. Additional complexities caused by environmentally conscious customers and the government initiatives to reduce emissions by pushing for vehicle electrification didn’t help. While in the short term, hybrid vehicles will continue to dominate; it seems clear that the industry is moving towards purely electric models. Electric vehicle (EV) manufacturing will impact OEM assembly plants and the Tier 1 and 2 suppliers.

All of these factors have accelerated the digital transformation journey amongst automotive manufacturing OEMs, Tier 1 and Tier 2 suppliers. Industry 4.0 and Industrial IoT will allow machine builders and integrators to equip their manufacturing lines with low-cost sensors that stream data to the enterprise or cloud for analysis. This analysis enables the tracking of the performance of operations right down to the component level. Combining this data with a digital twin of the factories allows operations to build an accurate representation of the process that helps problem resolution, continuous improvement, remote visualization, flexible planning, and change management. These would also enable global visibility into supply chains and electrification.

Over the years, the automotive manufacturing industry has invested heavily in enterprise systems, automation, and advanced product technologies. Automotive manufacturers are realizing ROI on their investment and refining their manufacturing processes by beginning to embrace IIoT. In addition, they are starting to measure the success of digitization initiatives which validates their efforts and helps them see the ROI of their investments. This whitepaper explores how IIoT in general and more specifically MQTT, Sparkplug are enabling the automotive manufacturing industry to help improve their worldwide operational efficiency, reduce supply chain disruptions and enable their push towards electrification.

How IIoT is Improving Automotive Manufacturing

The automotive industry, including manufacturing, is experiencing heightened regulatory scrutiny both in the US and globally, especially with the added complexities of electrification. This scrutiny is occurring in several areas, including quality control activities around manufacturing operations, materials used for manufacturing, and driver safety.

IIoT technology is making it possible to automate Automotive Manufacturing operations, optimize supply chains and enable digitized quality management. With the help of smart sensors and advanced gateways, companies can easily monitor the performance of a system in real time, automatically adjust parameters for optimum industrial efficiency, and ensure better productivity at all stages.

Some of the ways IIoT is benefitting Automotive Manufacturing include:

1. Getting access to real-time visibility of operations throughout the manufacturing process right from supply chain to production to distribution is key especially with trends like electrification, and lighter auto body materials becoming prevalent.

2. Allowing for modular automation in a manufacturing plant, helping to scale automotive production per market demands. This is important as the need to make cheaper, fuel-efficient, and feature-rich vehicles in different configurations continue to increase.

3. Identifying bottlenecks among several processes within a plant, while also indicating equipment utilization, which increases machine efficiency, reduces replacement costs, and increases productivity of the automotive plant.

4. Enabling faster, continuous, and feasible data collection from several processes in a manufacturing plant required for quality control and reporting purposes.

5. Powering the latest manufacturing best practices like additive manufacturing, and mass customization through advanced data processing and analytics.

How MQTT Addresses Connectivity Challenges in Automotive Manufacturing

The first step towards successfully implementing IIoT in Automotive Manufacturing is to establish connections to manufacturing machines, processes, databases, applications and supplies achieving a return on investment (ROI) and deriving value from IIoT depends on one critical factor: data. Operational data powers use cases such as machine condition monitoring, improving OEE, and implementing preventative maintenance. Data connectivity and availability are often the primary challenges for IIoT deployments. As a result, we’ve seen increased adoption of two open standard messaging protocols that aim to solve data movement challenges and help manufacturers become data-driven - MQTT and Sparkplug.

MQTT is a lightweight, publish/subscribe messaging protocol ideal for connecting remote devices (e.g., the IIoT). MQTT has a small code footprint. This design allows data to move within a challenging communications environment with resource constraints or limited network bandwidth. MQTT has become the de facto standard for device messaging in IIoT communications, with millions of devices leveraging its capabilities.

More recently, Sparkplug has emerged as a new open specification that, when deployed in conjunction with MQTT, compounds business benefits by orders of magnitude in terms of cost savings, time-to-market, reduced complexity, and increased productivity. We will focus on these business benefits in this paper, highlighting the positive outcomes and ROI that result from an MQTT/Sparkplug architecture. While some general understanding of the technical aspects of these deployments will help, our goal in this paper is to share how adopting Sparkplug will positively impact your business and help you compete more effectively while reducing costs.

What is Sparkplug?

Sparkplug is an open-source software specification for increasing MQTT interoperability, specifically in smart manufacturing and Industrial IoT. The specification provides operational technology (OT) data with context for seamless integration with information technology (IT) in a bi-directional and interoperable way. In simpler terms, Sparkplug enables the normalization of data at the edge to the cloud or enterprise data hub so it can be used to create a local Unified Namespace, for machine learning or other IIoT applications.

MQTT effectively enables device-to-device connection to any device or system in an IIoT framework, regardless of type or model. However, MQTT messaging provides zero context into what information it’s sharing. The addition of Sparkplug provides industrial data with the context necessary for IIoT architectures to take specific types of action, depending on what the data requires. With MQTT and Sparkplug, any application or device that wants access to the data can subscribe. New data sources can become immediately discoverable to other system components, and these data sources can become a single source of truth upon which the entire system can act.

A Sparkplug architecture may look like Figure 2, with Sparkplug-enabled OT systems on one side producing data, a fully Sparkplug compliant and aware MQTT broker in the middle to move the data, and Sparkplug-enabled IT systems on the other side consuming the data. Data can flow bi-directionally and Sparkplug adds valuable context including an MQTT topic structure definition, MQTT state management and payload data definitions.

As a result, Sparkplug provides industrial organizations with a few key capabilities. First, it standardizes and defines OT data so all subscribers know how to utilize it without special programming or coding. This enables the “single source of truth” we referred to above that eliminates confusion for an entire organization’s IIoT architecture and allows all cloud and enterprise systems to utilize the data.

Second, Sparkplug enables organizations to connect devices to infrastructure rather than applications. By doing so, the specification paves the way for other systems to ingest data for their purposes. Third, Sparkplug lays the foundation for true, mission-critical IIoT applications with ultra-reliability and low-latency operations. Without these characteristics, true IIoT cannot exist. Finally, Sparkplug enables a real, interoperable “plug and play” IIoT environment. Not only can messages be easily distributed, but there are also common definitions for what these messages mean, which devices they represent, and how they should be treated.

Let’s look at how adopting Sparkplug can positively impact your business.

Avoid Vendor “Lock-In” via Open Standards

In a market that has been using the same strategy for control and command systems since 1978, the advent of OT-IT collaboration should bring a welcome change in flexibility for industrial organizations. For decades, hard coding of systems and applications has meant companies have had to rely on single vendors. These vendors could effectively “lock in” their customers because the risk, cost, and integration issues of moving prove far too onerous to most industrial companies.

With an open specification governed by a vendor-neutral non-profit organization (in the case of Sparkplug, the Eclipse Foundation), industrial organizations now have an easy, cost-effective means to migrate from hard-coded, specialized systems to a world of open standards. Companies can easily upgrade standardized systems or swap out for another solution.

Applying open specifications like Sparkplug and open-source software to the industrial manufacturing sector will lay the foundation for a dramatic shift from workloads using proprietary interfaces and running on fixed-function hardware to interoperable and portable workloads running on commercial off-the-shelf (COTS) hardware. This shift will result in broadly scalable solutions that generate standardized data that support developing and applying more advanced value-add solutions for the industrial sector.

A transition of this scope will require more than just the Sparkplug specification itself. It will also need a global ecosystem of vendors and companies collaborating for the ongoing evolution of the industrial sector. This process has already been implemented at Sparkplug’s inception, as it is an open-source project governed by the Eclipse Foundation, the world’s largest open-source software foundation focused on IIoT and Edge technologies. The Eclipse Foundation has decades of experience providing vendor-neutral governance while still attracting industry leaders and innovative startups alike to provide input and guidance to the evolution of Sparkplug and other technologies.

Bridge the Gap from OT to IT

At its heart, IIoT is about leveraging the lower costs, innovation, and flexibility of IT, while retaining the high reliability of OT systems. The concept is not new, but many past efforts to bridge the gap were confusing and had technical obstacles that negatively impacted the industrial network. Without the data context Sparkplug enables, IIoT architectures often miss the critical “last few feet’’ required to allow OT and IT systems to work together successfully.

Sparkplug changes the game entirely. Defining the industrial purpose and intent via context allows IT systems to easily “ingest” and “understand” OT data, something not possible before Sparkplug without intense, lengthy, and flawed coding exercises by the industrial organization. That means significantly added costs, provided the organization could find the developer talent to take on the project.

Once Sparkplug has lowered the barrier to connecting IT and OT systems, where they can communicate, IT systems can then act upon OT data for advanced analysis and modeling. The ability to create a feedback loop and act on OT data, in turn, provides all the benefits of IT that lead to more productive and lower-cost architectures. Some of these new capabilities include:

• Real-time data analysis: Sparkplug-enabled architectures enable granular, real-time analysis of the data generated by every system and device in an industrial setting.

• Digital twins: Smart sensor data can feed digital twin models to run simulations before actual devices are built and deployed.

• Application of AI and machine learning: Sparkplug makes realizing the power of AI to deliver real-time insights into operational efficiency a relatively simple task.

• Remote monitoring and predictive analysis: Data moved from OT to IT by MQTT and Sparkplug can be used for advanced analytics that provide insight into machine operations and even enable predictive maintenance.

• Health and safety: Sparkplug enables IIoT applications that contribute to safe and healthy work environments.

• Advanced wireless technologies: Several mobile technologies used to advance supply chain applications, such as 5G, LoRA, Wi-Fi7 are now possible via Sparkplug.

• Agile architectures: IIoT networks built on MQTT and Sparkplug give organizations amazing flexibility regarding suppliers and how they manage their resources.

Reduce System Complexity and Scale Easily

Another key benefit of Sparkplug-enabled systems is reducing the complexity of the overall system deployment, both from an ability to easily replicate the software required at each site and reduce the number of sites required to service the same area. Many data movement solutions are enabled with custom code to manipulate the data without the ability to scale. Sparkplug gives factories one open data movement structure where multiple products and systems are Sparkplug-enabled, and templates can be defined and configured. Organizations can add new devices or sites rapidly with auto-discovery and templates.

An automation specialist for a major oil and gas company shared how leveraging Sparkplug for a single business unit reduced complexity and thus lowered software programming costs saving $1.3 million annually. The business unit was building 50 new facilities a year. By streamlining system architecture development for each facility, they achieved fiscal savings and sped up time-to-market.

The fiscal savings, considered a conservative estimate by this specialist’s accounting, is primarily due to Sparkplug enabling a “write once, use anywhere” model that sped time-to-value for building system architecture. His Sparkplug-enabled system reduced developers’ time to incorporate a new object from 30 minutes to 5 minutes per object (tanks, pumps, etc.) with MQTT Sparkplug, or a 6x faster integration time. Without MQTT Sparkplug, the firm needed to pay system integrators to build the same object as many as three separate times (POC, PLC, and HMI level), effectively re-inventing the wheel every time.

Even more dollars were saved by creating an MQTT/Sparkplug-enabled architecture that could easily observe system behavior in real time without having to send human beings out to remote sites. Instead, leveraging Sparkplug’s capabilities, his team could instantly call up all pumps or all tanks and determine their status on demand. These same systems could also proactively alert teams if there was an existing issue, using predictive analysis to identify if an event may be coming soon.

Beyond reducing programming time, Sparkplug systems can also reduce the number of 1:1 relationships in an overall system, lowering the amount of work to maintain dozens of sites simultaneously. Overall, this same automation and IIoT specialist reduced total costs at more than 50 sites by more than $2.6M annually.

Enable Mission-Critical Systems

Sparkplug is critical in enabling mission-critical OT systems that have a low-latency architecture. Sparkplug’s interoperable “plug and play” capabilities and its ability to serve as a single source of truth makes sophisticated edge computing architectures a reality. With edge computing comes distributed intelligence, reduced latency, and, therefore, greater scale using far fewer resources.

By definition, edge computing puts data and computing intelligence closer to the physical objects with which it interacts. This inherently reduces the latency in the system allowing for much faster reactions and improved performance. On the factory floor, a Sparkplug-enabled architecture can process terabytes of data more quickly and at less cost than a legacy OT system. From operations to predictive maintenance and more, this architecture enables businesses to take more rapid action than if they had to batch this information and send it offsite to be processed elsewhere.

Conclusion

The use of Sparkplug is proliferating across a myriad of industries. Major automotive, manufacturing, industrial, oil and gas, and supply chain/logistics firms are leveraging its capabilities today. Cloud hyperscalers are actively leveraging this specification to build new managed services targeting industry markets. This implies that if your competition is not already using Sparkplug they are likely to be actively evaluating it to integrate it in their environment in the near future. A recent survey from IIoT World asked companies building IIoT systems which data movement tools they consider essential to fulfill their IIoT strategy. MQTT was the most popular answer at 55%, and an impressive 25% answered MQTT Sparkplug, solidifying the relatively new technology as a key piece of the puzzle as companies digitally transform and roll out IIoT projects.

With a robust and open ecosystem supporting this technology’s development and widespread support for its deployment in the market, Sparkplug should be an essential part of your future IIoT deployment plans. Given its benefits, Sparkplug can and will give your organization a competitive edge.

When you’re ready to reap these benefits for increased ROI from your IIoT deployments, contact us to learn more about the fully Sparkplug compliant and aware HiveMQ MQTT broker.

Or, if you want to learn more about Sparkplug’s benefits, architecture, and how to implement it into your environment, read the Sparkplug Essentials Guide.