Information integration and IIoT Webcast: Questions answered
Information integration ensures that the right information gets to the right persons, automated systems, and devices in time to make the right decisions. Additional answers are provided below beyond what was discussed during the Dec. 3 Webcast (archived for one year).
Answers to additional audience questions cover challenges, early adopters, benefits, security, wireless, Ethernet, and infrastructure. Webcast includes an exam to earn a Professional Development Hour (PDH). The webcast also discusses Control Engineering research on "Industrial Internet of Things, Industry 4.0, and Information Integration."
The main Webcast speaker was Dennis Brandl, founder and chief consultant at BR&L Consulting Inc., specializing in helping companies use Manufacturing information technology (IT) to improve production, laboratory, and logistics processes. He is an active member of the ISA 95 Enterprise/Control System Integration committee, co-author of the MESA B2MML standards, a member of the ISA 99 Industrial Cyber-Security Standards committee, former chairman of the ISA 88 Batch System control committee, contributor to the OPC Foundation and IEC 62541 standards, and has participated in the development of other industrial standards.
While a lot of IIoT is technology, "Without effective policies, procedures, and training, the effort required to keep a distributed control system running will quickly overwhelm most control departments," Brandl noted. His answers on IIoT and information integration are organized into four areas: infrastructure, adoption and benefits, security, and networks and information integration.
1. IIoT infrastructure issues
Question: What is the biggest challenge regarding the IIoT implementation?
Dennis Brandl: The biggest challenge is that the infrastructure needed is not in place. Companies will need to expand their wired network capability, with 1 GB to 10 GB network backbones just for the production environments, and enable the use of secure wireless devices. Integrating these changes into a production facility can take nine months to a year, and many companies are still discussing the viability of wireless and the need for high speed production networks.
Q: Can you talk about roles of decision makers within the facilities? For example, who is responsible for funding these integration projects, or who is implementing security zones, or who is most interested in the analytics (operations or management)? Is there organizational restructuring taking place to implement these changes?
Brandl: As industrial devices move to use the same underlying technology of the business IT systems, there are increasing conflicts between the needs of engineering and production and the business back office IT systems. Generally, the engineering/production personnel define the security zones because they understand the operational risks.
However, funding of integration projects is usually funded by the business back office side, because they need the data from the factory floor to run the business.
Analytics is performed by engineering and production, usually as part of a lean or six-sigma project.
The most promising organizational structure I have seen is a specialized group within the IT department that focuses on engineering/production systems and has dotted-line responsibility to production management (vice president or director of manufacturing). The IT role is a support organization, but the specialized skills needed for engineering/production systems means that not every IT staff person should be allowed to manage the engineering/production systems.
2. IIoT adoption and benefits
Q: Who are "early adopters" of IoT concepts so far?
Brandl: Right now the early adaptors of IIoT devices are in the vendor labs. A few companies, looking at the 5-year and 10-year time horizon, are starting upgrades to their network infrastructures to handle the tens of thousands of devices that will be connected.
Q: How would IIoT benefit a company in a developing country?
Brandl: Developing countries will probably be the first to implement the distributed industrial control-enabled (DICE) devices, because they do not have the investment in current automated systems using programmable logic controllers (PLCs) and distributed control systems (DCSs). As DICE devices become commonly available, they will offer the opportunity for incremental automation projects and much reduced startup costs, helping developing nations.
Q: What is the accuracy in recorded percentage for technology change workflow?
Brandl: The move from manual workflows to automated workflows has been well documented in the implementations of the ISA 88 standards. Companies have consistently reported significant (over 50%) reductions in product variability and production time variability by using automated workflow systems. As more ISA Level 3 workflows are automated, there should be corresponding increases in nonbatch production facilities. The biggest source of errors in most production is human error, so the biggest improvements that can be made are to reduce the inadvertent and accidental errors from manual operations.
Q: How much does what came before IIoT have to do with the success of IIoT?
Brandl: The industrial standards that will be the foundation of IIoT and DICE devices are all based on 40+ years of work that has gone on before. It will be based on the well-developed standards for automation control logic structures, industrial grade security, safety, and information management. The opportunities provided by IIoT technologies finally allow us to apply our 40-plus years of knowledge into very small distributed control devices.
Q: Will all this be easier as facilities and manufacturers update their devices and systems? Or is it more than only technology?
Brandl: Entering the IIoT world is more than just technology. The policies, procedures, training, and general knowledge needed to manage tens of thousands of connected smart devices are as important as the underlying technology. Without effective policies, procedures, and training the effort required to keep a distributed control system running will quickly overwhelm most control departments.
3. IIoT security
Q: What measures are developing to prevent catastrophic data security breaches that would shut down operations?
Brandl: Everyone involved in the development of IIoT devices is well aware of the need for robust and reliable security. Most medium and large companies are also aware of the need for security. These companies will drive the use of standards, such as IEC 62443, to provide the secure environments needed in the IIoT world.
Q: What standard or standards are going to be adopted to reassure end users and customers about security? Who is working on that now?
Brandl: There are currently thousands of engineers and scientists working on the security standards needed in the industrial community. ISA is working with IEC in developing the ISA/IEC 62443 standards (Security for Industrial Automation and Control Systems) and has several standards and Technical Reports approved and available. This work is being coordinated with the ISO/IEC 27019 cyber security standard for process control systems and the ISO/IEC 27002 Information Technology Security Techniques. But the real proof of security will be through certification of professionals, companies, and products, such as the ISA Security Compliance Institute (ISCI) supporting the ISASecure Industrial Control Systems Cybersecurity Certification program.
Q: Can the IIoT devices be scanned for inventory purposes and identify vulnerabilities?
Brandl: We don’t have all the standards defined yet for industrial device management, but work has started in the ISA 108 Intelligent Device Management committee to address this issue and other remote device management issues.
4. IIoT networks and information integration
Q: How are DICE devices going to talk to each other? Are different wireless protocols going to be used, such as ZigBee, Z Wave, and others?
Brandl: In my opinion, the most commonly used protocols will be based on Ethernet running an industrial protocol on top of TCP/IP or UDP. Protocols such as OPC-UA, EtherNet/IP, and Profinet will all be used. Connecting to very small devices will probably be wireless and will use multiple technologies. Industrial devices (such as those used to run a production line) will need robust, secure, and noise resilient technologies. I believe that these will be WiFi and ISA 100 devices. The cost differential between ZigBee, Z Wave, etc., and consumer (WiFi) will decrease over time, and the ability to use one device management system and a distributed control environment will, in my opinion, drive the use of WiFi and related protocols even on low-cost devices.
Q. Are there any issues with wireless "noise" in an industrial environment with devices?
Brandl: In multiple studies the wireless "noise" has been shown to be manageable through technologies such as spread spectrum wireless. Standard WiFi and protocols such as the ISA 100 standard, use direct sequence spread spectrum to minimize the impact of industrial device noise. A lot of the radio noise is in a limited spectrum, and newer WiFi spectrums often allow the use of less noisy spectrum segments. Also placement of wireless hubs that connect to wired networks, proper antenna cabling, and antenna placement have been effectively used in high noise environments. Proper cables and antenna placement can usually reduce the impact of noise 100 fold.
Q: How about MTConnect and B2MML mapping?
Brandl: There hasn’t been any work in that area yet that I’m aware of. Usually this work is done through volunteers, and none have stepped forward yet. (Would you like to help?)
Q: Does MTConnect tie in with this at all, or is it a competing standard?
Brandl: MTConnect will be part of the solution. There will probably be no single protocol that will be used to communicate to all IIoT devices, but a multitude of protocols will exist for specific purposes. I am encouraged by the joint work on the OPC Foundation and the MTConnect developers to develop interoperability standards, and I can envision OPC-UA as becoming a container standard for interoperability, so that MTConnect and other industry specific protocols can interoperate.
Q: Where is the constriction of information flow the greatest and what can be done to improve it?
Brandl: In my opinion the biggest constriction of information flows will not be between the business and production systems, but within the production systems as huge amounts of data are used for analysis. A single-point limit switch, scanned every few seconds, can generate hundreds of gigabytes of information in a year. Very little of that information needs to flow to business systems, but all of it may be needed for operational analysis. Production and operations networks will need high-speed networks, far beyond the current 10 MB and 100 MB networks currently used. They will also need terabyte databases for analysis and improvement projects. Companies should start planning to upgrade their infrastructures now, because it usually takes multiple business cycles to get approvals and validated installs.
– Edited by Mark T. Hoske, content manager, CFE Media, Control Engineering, firstname.lastname@example.org.
Watch the related Webcast to see more on this topic.
See the related research study on the same topic, giving full results excerpted in the Webcast.
See the 2015 IIoT series of four Webcasts, an overview with advice and parts 1-3.
See more about ISA and IEC standards referenced above.