Designing a converged IT and OT ecosystem in a world of competing platforms, software, sensors and services
The convergence between information technology (IT) and operational technology (OT) creates a digital ecosystem of connected smart devices, sensors, software and platforms. The ecosystem provides opportunities in the energy and chemicals sector to create new connections that enhance capabilities and services that will improve customer satisfaction and performance. Organizations are wondering how to exploit the converged digital technology – some take a “wait and see” approach, some take the lead and fully embrace the opportunities, most are merely experimenting.
There is a vast array of digital technologies in the market as vendors compete for first mover advantages, with no sign of consolidation anytime soon. At the same time, advances in cloud and mobility have made it easier for organizations to introduce these new technologies and roll them out to staff. However, many of these are not yet mature: they work well at individual and team levels, but are not yet proven for organization-wide operation. The industry is trying to keep up by developing standards and protocols to ensure interoperability of the technology components, but is lagging behind the pace of technological development.
The new digital platforms, software, sensors and devices also blur the line between information and operational technologies. In the past, OT solutions typically comprised proprietary hardware and software products with their own proprietary protocols in a closed environment. Some of these protocols are now adopted as the standards, for example IEC 61850 is now the de facto communication protocol for electrical substation automation systems. Many organizations are gradually implementing IEC 61850 and vendors are developing IEC 61850 compliant products. Alarms and events generated by OT systems in proprietary formats are now digitalised and available as standardized and meaningful data that can be transmitted over IP-based networks. As a result, OT organizations must respond to emerging demands, such as storage for big data, cloud infrastructure, cyber security, and analytics, capabilities that have traditionally sat within the IT domain.
On the other side, IT is faced with the challenge of managing an increased volume of data and frequency generated in real-time by OT devices and sensors instead of the familiar human-generated data. Although the cost of infrastructure and data storage has significantly reduced over the last 10 years, the increase in data traffic has impacted management of infrastructure, storage and network performance. With OT being virtualized and connected to IT platforms and applications through open communications networks, it creates advanced capabilities in remote control and process automation. There are significant benefits to be realized through these capabilities if OT and IT organizations can work together to manage the risks and secure the converged technology ecosystem from cyber-attacks such as corporate breaches to steal information (IT) or unauthorized controls over a machine or a plant (OT).
The technology convergence is here and organizations need to respond. What should we use the new technology for? Which technology should we invest in, given the immense options and lack of maturity? How do we keep up with the pace of technological advancement? How do we know if we are making the right technology decisions? And how will the organization need to change and integrate to be able to manage these sustainably? These are some of the challenges that need to be answered.
Figure: Technology convergence across Operations Technology and Information Technology
What should we use the new technology for? Which technology should we invest in, given the immense options and lack of maturity?
Often, organizations can get too hung up on the technology instead of the outcomes it can deliver for the business and customers. Instead of going straight into technology selection, the business needs to first identify business requirements and use cases that can be enabled by IT/OT convergence. These use cases then need to be translated into technology capabilities to find patterns and common requirements.
The diagram below is a (very!) simplified representation of a common converged IT/OT stack. One way to discover the patterns is by categorizing the requirements by components in the solution stack. For example, when a machine fault occurs on the field, an operator will contact an engineer providing him/her with the fault details for the fault to be rectified. A use case is then identified for an automated machine fault notification to be sent to an engineer’s mobile app, allowing for a timely response. Building on the initial idea, the use case can be extended to enable the engineer to initiate shutdown from his/her mobile device depending on the type of fault to secure the machine.
Figure: Automated machine fault notification use case
We believe this process is useful for organizations to scope and prioritize use cases that will deliver the most benefits, balanced with the risk of failure. In the example above, a remote machine shutdown might bring more benefits than the automated machine fault notification. However, it also has more failure and production risks. Breaking down the use cases into digestible business requirements provides useful information to assist investment decisions. In this example, the decision to trial the automated notification use case first makes sense as it is the building block to do the subsequent use cases.
Having a structure (albeit a simple one like the table above) helps modularize the solution and narrow the scope on technology options while still being mindful of the ultimate goals. Most Energy and Chemical organizations already have the smart devices hardware and software in place with some connectivity back to a platform. To implement the automated fault notification use case, the organization can focus on finding a solution to store the fault event and perform necessary processing of the raw data for it to be presentable to a human.
Depending on the organization’s risk appetite and the sophistication of the use case, there might be a need to trial new unproven technology. Vendor partnerships are critical in this scenario where both are co-designing the new solution. This partnership can be successful provided the organization is clear on the risk exposure and able to contain the cost and risk of failure. This of course needs to be balanced with potential benefits and creation of new capability.
In the case of implementing IEC 61850, the cashable benefits are not clear upfront. However, as it becomes an industry standard, integration costs will become lower and new capabilities will be enabled through new, compliant hardware and software.
How do we keep up with the pace of technological advancement? How do we know if we are making the right technology decisions?
After making careful considerations on which digital technology to invest in, people can often get too attached with the chosen technology. Regardless of the use case, flexibility and scalability should be key features of the solution, especially as standards continue to evolve. This allows organizations to switch components of the solution quickly when things don’t work and build new functionality when things do work. It is important to stay committed to the idea and the decisions. However, organizations should continue to challenge the solutions that they put in place and keep their eyes open to what’s happening outside the organization.
The new connected world will not be built on a straight line. While there is a lot of external focus at the beginning with market scans, peer reviews and vendor selection; it is important to start turning attention internally to ensure a successful technology implementation (or a fast failure). The major milestones in this cyclical process are illustrated in the diagram below. The process typically starts with an idea or a belief. Someone had an idea that if we can predict when a machine will fail before it fails, we will be able to proactively plan for the downtime required to conduct the preventive maintenance with no or minimal impact to production throughout.
Figure: Fast failure process
After breaking down the predictive machine fault alert use case using a framework like the one discussed above, having a data lake and rich analytics solution is a logical first step. There are many cloud solutions available for this which means the cost of failure is relatively low, i.e. minimal capital investment with the ability to turn off the solution at any time. If the cost of failure is high, more due diligence might be required to prove the business case.
In a world of competing platforms, software, sensors and services; everyone has their own preference when it comes to technology. Understanding the organization’s commitment is critical to decide the initial scale of implementation as the success of any technology is only as good as its level of user adoption. For the predictive machine fault use case, sponsorship from production manager and maintenance engineer is crucial to prove business value from the solution.
The initial solutions might involve developing some algorithm to process historical fault events combined with information about plant operation and the machine itself. This will become an iterative process whereas value is demonstrated, the organization increases its commitment. The greater the organization’s commitment, the greater investment to either add functionality, adopt better technology and/or to scale to other plants. This demonstrated value validates the initial belief and generates new ideas – i.e. Can we get the machine to automatically rectify the problem for simple failure reasons? Can we get the applications to automatically order new spare parts? When should we move to streaming analytics for real-time control? etc.
Rise above the fray
Technology convergence is here but organizational integration is not. It is important to understand what the convergence means for the organization and then work out how to integrate within and across technology, data, process and people, with security front-and-center. New technology comes and goes. Having an all-inclusive architecture across IT and OT components makes the technology decision-making process simpler and is focused on the intended business outcomes.
The new digital technology and its connectedness is moving faster than the industry’s ability to define standards, and much faster than a traditional organization’s ability to adapt to the changes. The big data brought about by the convergence is meaningless until organizations can create and adopt new ways of working to exploit it. This requires a new type of organization – one that constantly changes as it responds to the insights produced by technology.
This text comes from Cordence Worldwide’s new collaborative publication “Leading Successful Digital Transformation across the Energy and Chemicals Sector.” Please visit the CWW website to read and download the full perspective.