Written by
Sonia Bot, Sheppard Narkier and David Sherr
This is Part 2 of a three-part series based upon our new book, Dynamic Multi-Level Decisioning Architecture: Making the Right Decisions, at the Right Time, with the Right Authority for Sustained Competitiveness and Relevance. It’s about how all industries, including the rail industry, its partners and adjacent competitors, are evolving in the throes of digital disruption and other external forces, and emphasizes the role all employees play in decision-making at all levels of an organization. Decision-making needs to flow better upwards and downwards, tying the boardroom to the railroad yards, transloading facilities, tracks—the whole rail network and transportation ecosystem.
The Business Environment as a Service (BEaaS) platform can be viewed as enacting the end product of our decisioning architecture. It’s what stakeholders and users can see, touch and experience—a natural digital meeting place for buyers and sellers of products and services. What makes this environment special is that stakeholders can be on any side of the market. It will make sense for some niche players to be providers to a larger provider who consumes a customized offering to provide a richer service that neither provider could offer by themselves.
An example is transloading, where several rail freight ecosystem participants such as rail, trucking and marine operators; shippers (senders and receivers); and trade financiers. The ecosystem comes together in a neutral digital meeting place, all with the incentive and accountability to make sure they benefit from smooth operations.
BEaaS is a platform that supports onboarding, updating, discovery and orchestration of sets of industry-related services, which are mostly supplied by ecosystem partners. It creates a marketplace where newer services are less expensive to start up and maintain because the basic digital infrastructure is in place. The platform enables services to be changed faster and more frequently, a necessity in an age of constant disruption.
A good example is a dynamic railcar allocation model using a variable number of cars for specific roles based upon real-time assessment of incoming loads at key transfer points. This could greatly improve fuel efficiency and make better use of car utilization. It should enable railroads to be more quickly responsive to unexpected events with better flexibility.
The foundation of any IT service is provided through messaging, data delivery, service discovery and maintenance. The goal is to easily integrate these services from different providers to deliver value faster with less startup investment exposure. In the case of any type of machine learning such as for digital twins, it is possible to have statistically aligned synthetic data available for secure model test runs, allowing for faster modeling and experimentation.
Another example would be railcar and locomotive maintenance in which, because of real-time feedback, a set of services and parts were projected in real-time to a service center in advance of train arrival. This would save time and improve service. A platform like this has enormous competitive advantages.
Things wear out in the universe even without use. Nicola Tesla’s insight painted a big picture when he said, “If you want to find a secret of the universe, think in terms of energy, frequency and vibration.” Energy, frequency and vibration put substantial stress on moving parts of rail freight equipment and infrastructure. There are ancillary assets—signals and switches, power lines that will need period repair. And last but not least, labor. Timely and proper maintenance underpins the performance of PSR, which improves vastly in a smoothly running system. That’s why decisioning on maintenance is critical.
What is Decisioning?
Decisioning is becoming a mainstream business term. It is decision management for design automation, AI, machine learning and secure digital twin technologies. A decisioning architecture with the convergence of Operational Technology (OT) and Information Technology (IT) as the cornerstone enables making the right decision, at the right time, with the right authority and scope.
It starts with how people assess value, which is in the eyes of the beholder. Each person has a value lens based on their role, the implied scope of their decision-making and the risk/reward calculation they make. For example, a brakeman would ask, “Could I lose my job, my life, or be injured?” A CEO would ask, “Could I cause a massive operational failure at a critical time that devastates my company and my reputation?” Both have very legitimate concerns. The difference centers on the impact scope and timeframe.
Another key aspect is the time in which a decision must be made. A brakeman might have less than a minute, a railroad manager less than five minutes, and a logistics dispatcher a day. The C-suite might need less than a month for a critical matter. Also, we need to consider the impact scope if a decision is not made. On a railroad, it could be massive. The principle is to limit the downside of not acting. A decisioning architecture defines meaningful components at certain levels of detail, then describes how these components interact with each other.
OT-IT Convergence
OT is often referred to as the “non-carpeted areas” of IT. Both involve computing, networking and storage technologies. OT is a category of computing and communication systems to manage, monitor and control operations with a focus on the physical devices and the operating processes they use. IT focuses on business support and business enablement by using technology to collect, manipulate, analyze, and generate insights from data.
In its most basic form, OT-IT convergence connects OT and IT systems, allowing them to share data. The goal is to use this connectivity to enhance the value these systems deliver. For example, consider an IT system connecting to a shipping container’s OT system for perishables. When the temperature crosses a buffer threshold, an alert is sent, and proper corrective action can be taken before the shipment spoils.
OT-IT convergence is driven by top-level business objectives, which are continuously managed and governed under an overarching executive-level authority. Common processes are the foundation engine while standard technologies and people competencies are the fuel. The payload is the knowledge that travels through the processes in the two worlds of OT and IT. And this knowledge can be refined to uncover latent needs and business value. The software connects, transports and shares the data throughout the converged system. IoT (Internet of Things) plays a substantial role.
Secure Digital Twins
A digital twin is an intelligent agent, a virtual representation that serves as the real-time digital counterpart of a physical object or process. We build them with intelligent agent tools. Digital twins mediate data and information between OT and the counterpart, IT; ergo the convergence happens. They are built by gathering data and information about anything you want to make it a copy of, and then recreating it in a digital space.
As an intelligent agent, a digital twin is out of sight, but certainly not unfelt. They are centrally situated in our decisioning architecture. The digital twin uses AI and machine learning, mostly to simulate the effects that change in a design process or conditions would have, all without subjecting the real-world object to those same changes. For example, suppose a management team is wondering if changing the maintenance schedule of the fleet of railcars and locomotives as well as other vehicles would positively or negatively affect the delivery of freight at their sites in the collective supply chain.
The same can be said of passenger routes and running these trains close to schedule for a great rider experience. If a digital twin of these operations is in place, then simply change the schedule there and find out without necessarily disrupting and putting the ongoing operations at risk.
The use cases in our book explore the example of how secure digital twins can assist in extracting value from supply chain business operations by being the nexus of OT-IT convergence for the physical asset. We start with key business processes, drilling down to keeping assets in working order so that goods and processes flow smoothly and efficiently.
Real-time or near real-time input is becoming a greater part of managerial understanding that will change approaches to operational tactics and overall strategy. Digital disruption and IoT are driving this change. Inexpensive, user-friendly sensors and advances in network access are making this possible. This technology is pervasive and will intentionally remain low-cost due to economies of scale. But they are more susceptible to hacking by entities outside firewalls.
Low-cost sensors, distributed data and diverse network traffic provides opportunities for system compromises at all levels of an organization. We need to design around this constraint, and this is where secure digital twins come into the picture. Secure digital twins have several layers of built-in protection. Data is scrubbed well before it is placed in repositories and has strong role-based access controls. Since sensors are devices that will have a simple digital twin model, sensor data needs to be carefully orchestrated.
Secure digital twins provide levels of predictability regarding how an operating device will respond to varying stress points from simple wear and tear to a significant breakdown. They can predict the need for maintenance well before the physical component fails. They can predict wear and tear impact on the operating mode of the entire machine, in real-time. If structured properly, they can work with an operations dashboard to alert of impending changes or even send the data to an operational control component to change an operating mode in real-time. This move toward greater autonomy is a game changer in terms of safer, more effective, and more efficient operations.
The BEaaS has decisioning architecture to support decision-maker actions. It’s an architecture with tools to develop multi-tenant SaaS (Software as a Service) instances that are consistent with that architecture. And it’s interoperable with any IaaS (Infrastructure as a Service) offerings. Security and privacy are table stakes for a BEaaS platform.
The BEaaS Platform
As OT converges with IT, the digital twin is the IT connection to the OT device. It’s an actual image of an intelligent agent that not only discovers insights but also can act upon them (autonomy). It continually updates the state of the device through the stream of sensor data, exclusively mediating messages among IT services. The OT device is one of the very few cases where something is hard-coded due to sensor limitations. The digital twin notifies a device of state change requests and contains and runs analytic measures of the device’s behavior.
The secure digital twin operates in a “cloaked network,” providing digital and some physical security and privacy levels. BEaaS security digitally certifies private and secured digital twins. As well, the digital twin“knows” any instruments used by the device to promote physical security. This is called physical surveillance. Regulators dictate operational safety surveillance. Operators act, regulators surveil. These regulators can be inside the enterprise and/or outside the enterprise, in government agencies like OSHA, FRA, NHTSA or nonprofit industry groups.
People typically make investment decisions based on incomplete and aged information, while market, technological, economic, regulatory, weather, environmental and other forces and events around us continue to change, at faster rates. Ambiguity tends to increase. The continuous influence of external forces adds another layer of complexity. We find that most decision-makers don’t really understand the technical debt and overall enterprise drag that already exists in their span of influence and control, especially in systems with an abundance of legacy, whether it’s technology, processes, skillsets, policies, regulations, products, or services.
As such, investment decisions are made all over the map—doing nothing (continuing as is), tinkering (spending effort and resources on low or no return or irrelevant initiatives), “Big Bang” transformations filled with hope, tiptoeing forward by underfunding, etc. Often, the most effective “win-win-win” options are the most powerful and sustainable options, but they were not taken because they were not properly discovered. Conversely, the proper option vetting ended up being improperly biased. Enterprise drag and technical debt keep piling on. Yesterday’s development is today’s roll-out and tomorrow’s legacy and debt.
Our decisioning architecture has traceability and visibility, with all the dots connecting. Once a set of uncontrolled chaos forces are discovered, a closed loop—a control system with a feedback loop—activates. This presents the opportunity to control the chaos. Organizations can assign varying levels of authority to effectively execute tasks that sustain competitiveness. Staged investments in these new technologies can be strategically introduced so that subsequent phases can be paid for by savings obtained from prior phases.
Next month, Part 3 will cover success factors that lead to the sustainable evolution of a decisioning architecture.
Sonia Bot, chief executive of The BOT Consulting Group Inc., has played key roles in the inception and delivery of several strategic businesses and transformations in technology, media and telecommunications companies worldwide. By utilizing methodologies in entrepreneurship, business precision, Lean Six Sigma, system and process engineering, and organizational behavior, she’s enabled organizations to deliver breakthrough results along with providing them a foundation to continue to excel. Sonia’s contributions to the rail industry are as a leader and a visionary who is passionate in taking railroading into the next generation. Within the Digital Business Transformation context, she leads high-stakes mandates where new business models are created and enabled by digital technologies. She was instrumental in PTC implementation on CN’s U.S. lines. Her approaches on the evolution of railroading and transportation are game-changers that drive innovation and competitive advantage for adopters in a changing industry. Sonia can be reached at [email protected] or www.botgroupinc.com.
Sheppard Narkier, a CTO and Senior Enterprise Architect, is routinely tasked with complex, difficult transformation projects in a range of industries including demanding Capital Markets environments. Sheppard was recruited into 11 unique roles including Chief Technical Architect at a global investment bank, CTO of a global consultancy, and Chief Scientist at a startup. He has defined the architecture and rules systems for several application and infrastructure design platforms resulting in seven awarded patents. As a facilitator of change, he has driven the organizational transformations aligning systems development structures, processes, and data repositories with their strategic goals. A pioneer in cloud strategy, he developed IP in several companies to guide enterprises toward staged migration to hybrid multi-cloud across a range of horizontal and vertical scenarios. He has also employed multidisciplinary Design Thinking in recent engagements. Sheppard can be reached at [email protected].
David Sherr is in his sixth decade as a practitioner, thought leader, and executive in entrepreneurship, system design and development, and enterprise architecture. He has worked in six world-class financial institutions where he held CTO and VP positions and consulted in technology strategy for Fortune 100 companies. Currently, David is heading an IoT startup to build predictive maintenance analytics for industrial assets. As an inventor, he is named on six patents covering enterprise data management, Web services architecture, IoT Digital Twins, and, most recently, software-designed network resource provisioning architecture. David can be reached at [email protected] or www.newglobalenterprises.net/SCA.