How integrators can help IIoT applications
End users may find the variety of IIoT devices and connectivity technology to be intimidating. A system integrator can minimize risks and costs by implementing proven specification and implementation methods and architectures.
- The Industrial Internet of Things (IIoT) has reached the point where it is easy to find commercially-available solutions for plant-floor operations.¬†
- They‚Äôre not always easy to implement, though, and a system integrator can help with the process.¬†
- Technology has improved to where remote monitoring solutions can be developed in hours rather than weeks.¬†
By now, many commercial and industrial end users are well aware¬†of¬†Internet of¬†Things (IoT)¬†concepts, at least in general terms.¬†They know smart devices and sensors¬†can¬†communicate¬†with each other, the¬†internet, and mobile devices.¬†Many are interested in taking advantage of the¬†easy and inexpensive¬†data¬†access¬†promised by the IoT, but they¬†don‚Äôt know where to start.¬†Or, even if they have some good ideas, they are reluctant to dedicate personnel to the task¬†with¬†other pressing business priorities.
Many¬†good reasons¬†causer¬†trepidation.¬†For example,¬†common consumer-grade IoT gadgets¬†may be fine for home entertainment¬†systems, but¬†most¬†lack the¬†robustness needed for business-grade users.¬†Industrial IoT (IIoT)¬†devices¬†are available and¬†offer¬†improved reliability¬†when¬†they‚Äôre properly¬†selected and implemented.
IIoT networking, software, services
IIoT¬†projects¬†encompass¬†a range of¬†field-located hardware devices, networking, software, communication protocols, cloud services,¬†user devices¬†and more.¬†Just because any one device or software application claims to be¬†IIoT-capable or -ready does not mean it will be plug-and-play.
Some¬†automation¬†hardware, software and service¬†providers¬†have¬†embraced the uncertainty, researched the options, and developed¬†proven means and methods for delivering robust¬†IIoT¬†solutions¬†to¬†end users.¬†Multi-discipline¬†knowledge of industrial automation¬†hardware, electrical design,¬†networking, software programming, and cloud systems¬†are applied¬†to create¬†useful and reliable¬†ways of deploying¬†IIoT¬†applications.¬†End users can engage¬†an¬†IIoT¬†solutions provider¬†‚Äď¬†either directly or via a system integrator¬†(SI)¬†‚Äď¬†to create an economical¬†and effective¬†IIoT¬†installation, sometimes by filling out a spreadsheet.
Why IIoT?¬†Remote data access
Original equipment manufacturer (OEM) machine builders and¬†manufacturing/processing companies¬†already have a lot¬†to do when¬†designing and¬†operating¬†equipment¬†and systems¬†to¬†operate¬†productively.¬†Best performance is achieved when¬†the machines and equipment can be monitored, optimized, and diagnosed.¬†Even more efficiency is gained when users can remotely adjust¬†operational system parameters.¬†All¬†these tasks require¬†remote¬†access to the system data.
However, machines may be widely deployed throughout a facility,¬†and¬†at multiple locations¬†worldwide. Operational processes may rely on many remote¬†systems, such as pumping stations,¬†renewable power generation sites,¬†or tank farms.
Getting data from distributed or remote sites¬†has¬†often¬†proved problematic.¬†Establishing remote connections¬†has been¬†possible,¬†but¬†has¬†required¬†expensive¬†engineering¬†and installation efforts, including¬†radio or networking infrastructure. Cellular communications¬†improved things¬†somewhat,¬†but¬†it¬†could become expensive depending on the data quantity.¬†While¬†connectivity in place¬†problems¬†could be¬†detected,¬†it¬†was¬†often necessary¬†for employees to travel to site¬†to perform¬†additional diagnostics. This introduced¬†delays,¬†costs,¬†and¬†risks.
IIoT¬†implementations use¬†hardware devices, software applications, and networking options to¬†overcome these challenges by¬†making¬†it easy and cost-effective to achieve remote connectivity,¬†data monitoring, and adjustment,¬†but only¬†if¬†the end user can¬†specify, design, integrate, install, and maintain¬†the right combination of technologies.
IIoT good,¬†bad, and no¬†ugly
Some good news for¬†potential¬†IIoT¬†adopters is much of the technology has progressed to¬†a¬†commercial off-the-shelf (COTS) level.¬†Many sensors and field devices have¬†sufficient¬†built-in computing power. Wireless networking and cellular options¬†are fast and reliable.¬†Internet connectivity is ubiquitous,¬†and there are many¬†cloud computing choices. Everyone has a smart mobile device.
The bad news is that many OEMs and end users¬†are staffed to build and operate products¬†and processes, but¬†most¬†don‚Äôt have¬†a specialist¬†‚Äď¬†let alone multiple specialists¬†‚Äď¬†experienced with¬†IIoT¬†technologies such as¬†digital component¬†selection, software configuration, local area networking, internet/mobile networking, and more.¬†Even if they did have this staff, such a team might spend a good amount of time experimenting to develop¬†a solution.
Many¬†end users, as a result,¬†find¬†that engaging an¬†experienced¬†multi-disciplined¬†solutions¬†provider¬†is the best way to¬†cost-effectively¬†implement¬†an¬†IIoT¬†solution¬†and avoid¬†ugly challenges.
Building an IIoT Solution
Because¬†end users operate in a¬†variety of industries¬†using¬†a wide range of platforms,¬†there are countless technical variations possible.¬†Any¬†IIoT¬†solutions¬†provider must be in a¬†position to survey¬†the landscape of technologies and use cases to determine¬†the best¬†approaches.Original¬†equipment manufacturers¬†(OEMs), municipalities,¬†manufacturing/processing plants¬†and SIs¬†can benefit from an¬†IIoT¬†solution that includes remote monitoring and the following Figure 1 elements:
- Application data sources:¬†Typically instruments or smart systems in the field
- Edge computing:¬†A device to gather and pre-process the data
- Communications:¬†To transmit the data via a wired or wireless local network, internet connection, or cellular connection
- Cloud computing:¬†To aggregate, log, and¬†support visualization of the data
- Built-in and custom modules:¬†For measuring, calculating, and analyzing data¬†to provide insights necessary for informed¬†user decisions
- Mobile/web accessibility:¬†For users to view the resulting dashboards,¬†access data, and make adjustments.
Crafting an¬†IIoT¬†solution¬†begins¬†at the¬†industrial¬†‚Äúedge‚ÄĚ¬†where the data¬†is sourced¬†in the field.¬†Instead of forcing a one-size-fits-all answer,¬†there are a few¬†edge computing options¬†to provide¬†sufficient configuration or programming¬†capability¬†for¬†gathering¬†and pre-processing¬†data:
- Programmable logic controllers (PLCs): Primarily used for automation
- Human-machine interfaces (HMIs):¬†Primarily used for visualization
- Edge gateways:¬†Primarily used for data handling and computing.
The next step¬†is¬†enabling¬†data¬†transmission¬†to higher-level systems.¬†The edge computing hardware must have an¬†internet connection or mobile cellular data connection.¬†Also, due to the remote nature of these connections, any communication method must¬†accommodate outages, while¬†minimizing¬†bandwidth¬†usage¬†to reduce costs.
Communication protocols such as OPC¬†Unified Architecture (UA)¬†and HTTP are possible, but for this service the¬†message queuing telemetry transport¬†(MQTT)¬†protocol is ideal.¬†MQTT is a publish/subscribe¬†(pub/sub)¬†protocol, where edge-sourced data is only published to a server¬†when it is changed¬†or on a defined interval. Any clients can subscribe to the server to get the latest data when it becomes available. Also, MQTT¬†communications are initiated at the edge using outbound messages, which¬†generally means IT does not need to get involved with firewall configuration.¬†Depending on the PLC or HMI‚Äôs make and model, the device might¬†include¬†native,¬†but limited,¬†MQTT¬†pub/sub capability.¬†An¬†edge gateway¬†also¬†can provide this link¬†with more features.
The MQTT¬†broker¬†can be hosted on an on-premises server, but the more common and flexible method is¬†locating¬†it on a cloud-hosted server,¬†such as those offered by Amazon Web Services (AWS).
With these COTS elements in place, the next step requires developing software which processes the incoming MQTT data and delivers visualization dashboards to end user browsers and mobile devices. It should offer enough computing capability so users can analyze the data to create key performance indicators (KPIs) and other calculations supporting improved operations using preventive and predictive methods. The IIoT application also needs to be flexible enough to satisfy any client application, easy for an end user to work with and maintainable.
The best option for addressing these requirements is for a solutions provider to create a configurable cloud-hosted platform and offer it as software as a service (SaaS). End users may choose to access and configure the software themselves, but most will probably retain an SI‚Äôs services. for this work (Figure 2).
Another crucial aspect of any network or cloud solution is¬†cybersecurity.¬†SIs¬†often¬†will¬†enhance the built-in features¬†of the selected components and services to provide the following¬†protection methods:
- Each field device is identified with a unique ID and password¬†and connects using a WebSocket over SSL (WSS) tunnel
- One-way encryption using SSL/TLS certification and encryption¬†is the default, with two-way x.509 certification available.
- All servers and brokers are hosted on USA-based AWS¬†Cloud or AWS GovCloud, with¬†dedicated¬†server and portal deployment available
- Server infrastructure is based on horizontal scalable/repairable microservice clusters to provide uninterruptible service
- Application is designed and maintained by¬†a¬†NIST and GovCloud¬†certified and registered development team
- Internal audits are performed regularly
- Infrastructure¬†updates¬†and patches are applied¬†as needed
- Platform¬†can be audited by¬†any third¬†party
- Activities are monitored 24/7 to¬†look for breaches or attacks, and disaster recovery and security recovery¬†procedures and policies are in place to guide action as necessary.
- Data transactions between servers and user interfaces are secured by SSL/TLS encryption, with all certifications and encryption keys renewed/replaced every 60 days (most websites do this only every 360 days)
- User sessions are secured with complex passwords, and all logins and activities¬†are traced in compliance with NIST protocols
- A granular access control list (ACL)¬†ensures¬†each user¬†can access exactly as many or as few resources¬†as specified.
Anyone evaluating an¬†IIoT¬†solution should ensure the preceding¬†cybersecurity elements and best practices are in place.
Although this is based on many elements, it can be tailored to any given end user application. This is where many end users may want to partner with an experienced SI to achieve the desired results. To help streamline the process, a system integrator can provide a spreadsheet or other application where the end user can list the input/output points of interest and some supporting information. With this in hand, the SI can suggest one or more edge computing hardware configurations.
Once the concept is approved, the¬†integrator can:
- Design remote panels
- Specify edge computing and networking elements
- Configure¬†and¬†program the edge hardware
- Network the information to a cloud-based MQTT server
- Deploy¬†and configure¬†the¬†cloud¬†application
- Establish user accounts and dashboards.
Once the field hardware¬†is¬†installed, the¬†system integrator¬†can have a working¬†IIoT¬†remote monitoring solution up and running in hours.
IIoT in Action
One machine shop operates nearly 100 punch presses and computer numerical control (CNC) machines, each capable of supplying important data, which can be logged and analyzed to provide performance and preventive maintenance information. To monitor and collect this data, the SI designed and installed a remote terminal unit (RTU) panel at each machine. Each RTU used a PLC to monitor hardwired or serial data, transmitting it to another PLC acting as a supervisory data concentrator (Figure 3).
This¬†supervisory¬†PLC, in turn, was integrated with¬†the cloud-based¬†IIoT¬†solution¬†for logging data to a database, serving up dashboards with real time data,¬†making data downloads available,¬†and executing analytics rules to provide preventive maintenance reports.¬†Using this¬†available¬†information, the end user¬†was able to¬†increase machine availability, saving time and money.
In another case,¬†a highway transportation agency¬†operated many¬†stormwater pumping¬†stations¬†at road underpasses, although each¬†location¬†might have a somewhat unique configuration.¬†Initial automation and remote monitoring efforts were focused on providing¬†virtual private network (VPN)¬†remote access¬†for¬†power¬†and motor¬†monitoring, high level detection, seal leakage detection and weather sensing.
A later phase tied each of these subsystems together as a far more useful whole because all pump stations were integrated into the IIoT solution. Now, agency employees with the proper security credentials can use any web browser device such as a laptop, smartphone, or tablet to see the status of all the pump stations, and receive notifications if there‚Äôs trouble. This resulted in increased uptime and availability of all the stations, while reducing the chance of any hazardous road flooding conditions.
Technology has reached a tipping point where edge computing, networking, and cloud options are available for building IIoT remote monitoring solutions to help industrial and commercial OEMs and end users operate and optimize their systems. However, assembling the right mix of hardware and software still requires an experienced hand. Many users are not staffed for this or need to maintain a focus on their core business, which can be a barrier to adoption.
Pioneering¬†solutions¬†providers¬†have worked to create¬†a range of flexible¬†hardware architectures,¬†software solutions, and infrastructure¬†based on COTS¬†principles¬†and SaaS,¬†with a goal of quick and cost-effective deployment¬†backed up by ongoing support.¬†End users¬†can have a¬†workable and practical cloud-based¬†IIoT¬†remote monitoring solution up and running in hours, instead of weeks or months,¬†by¬†retaining an¬†SI to implement¬†and maintain¬†the solution.
Maryam Afshar¬†is a¬†product¬†manager¬†and¬†¬†Brian Gallogly¬†is president at¬†. Edited by Chris Vavra, associate editor,¬†Control Engineering, CFE Media and Technology,¬†email@example.com.
Keywords: IIoT, Industrial Internet of Things,¬†message queuing telemetry transport (MQTT)
What are your biggest challenges and concerns for implementing IIoT in¬†a¬†facility?