Commentary

Why “Data Soup” Provides No Real-Time Nutrition

Written by Daniel MacGregor, Nexxiot AG
image description

Vector flat global transportation concept illustration. MIT Sloan Management Review.

In my most recent Railway Age article entitled “U.S. Railroads Resurgent With Digitization at Just the Right Time,” I described how the railroads were the driver of U.S. growth, expansion and prosperity. They provided the spine for the economy to flourish.

The article went on to discuss some real-world use cases to demonstrate that by implementing the tools of digitization, rail can increase its role in value creation again. This article will bring some other use cases into focus and at the same time tackle the high-level perspective on the data side and discuss the prerequisites for meaningful analytics.

We can view the rail freight business at various levels. Components, Processes, Assets and Fleets are examples of this. Taking an asset perspective for now, we can see that the asset is on one level, an object and on another level a focal point for various events, business processes and customer and partner interactions. We can consider quick wins in maintenance intelligence and therefore maintenance and system performance, and we can consider improved services for cargo owners as immediate sources of new revenue.

What does it take to leverage such opportunities? By equipping non-powered mobile assets with low-power sensor nodes that harvest energy from the environment, it’s possible to get transparency and control. It sounds simple right? Well, it can be as long as some critical points are addressed from the outset.

  1. Near-real-time data collection and processing enables maximum extraction of value. Therefore, the solution must be highly redundant and zero maintenance/ “self-healing” and offer a frequent data update rate.
  2. The solution must be fully integrated for end-to-end functionality and offer interoperability with third- party sensors, platforms and control systems. This includes meeting all industry certifications and driving best practice on standards to enable operation in all industrial environments.
  3. The analytics must be developed in line with maximum “bang-for-buck” business objectives. The solution must be intrinsically configurable.
  4. Experienced, innovative vendors with a complete service suite are required as this is a consultative, iterative journey with a product that can be used to unlock each customer’s individual growth drivers and leverage value from cargo owners and, indeed, all participants in the ecosystem who can value the data or its extracted findings. The asset owners are in a privileged position—by default, with the right to equip the asset, and they sit on top of the value hierarchy to monetize the information as they wish.

Going on to consider the asset further, it’s worth raising the topic of the digital twin. For those unfamiliar with this concept, it can simply be described as follows: A digital twin is a simulation model built using real data gathered from its real-world physical twin, the asset itself. So, in the case of a railcar, the sensors send the data to the cloud; the data gets configured into a digital model of all the vibrations, impacts, loadings, couplings and interactions.

What’s the point of this? It becomes possible to re-imagine business processes. The digital twin can be used to project scenarios on the fleet to establish likely business impacts and outcomes. It enables business leaders to look into the future and say, “”What if we did this?” or “What if we altered our offering to absorb these new customer requirements?” In effect, you are creating an empirically based “forward perspective” on all decisions made today.

Let’s consider maintenance. Through the digital twin, we open up the possibility of true “condition-based maintenance.” With the location, the impacts, the vibration records and other sensor information, it’s possible to get valuable insights and clues into asset health. As data grows, analytics provide continual opportunities for immediate cost reduction when combined properly with classical domain knowledge of the industries’ operational veterans.

This is how effective digital twins are constructed. The long-term role of an integrated digital twin is to support all stakeholders during all lifecycle phases to support cumulative increases in productivity. Until now, design and operation have largely been disconnected lifecycle phases. I think there is a lot of instinctual brilliance in the rail industry. There are hunches that turn out to be true, and human intelligence and communication provides the practical solution finding that keeps operations live. These hunches can be given voice and opportunity for deep exploration with zero risk incurred with the right digital tools at hand.

Let’s take a few entry use-cases or “low-hanging fruit” as they might be called. We can gather anecdotal evidence on where to apply our inquisitive nature.

A railroad operator described this scenario to our Dallas-based solutions team. A Class I railroad locomotive was coupled to a high-value tamper that cost more than $1 million, hauling it  across the country at high speed, causing almost $100,000 in damages to the asset and infrastructure. When the case was investigated, some authorities denied it had even taken place at all. It’s a familiar problem to assign accountability where there is no evidence or data. When damages are incurred, the industry is beset with finger pointing and legal ambiguities.

Accurate, high-resolution data and the resulting legal culpability delivers a new level of respect for assets, processes, responsibilities and accountability, which in turn drives numerous benefits in administration, litigation and investigation.

In another incident, a Class I lost a boxcar full of supplies and equipment that was without an AEI tag. It therefore had no visibility, no CLM (customer lifecycle management) data and was entirely vulnerable to theft, misdirection and loss. It was missing for more than a week, and was eventually discovered in another city. This may initially seem fairly trivial, but an entire crew of more than 50 workers had a full week’s delay, which resulted in costs in the hundreds of thousands. Further costs were then incurred as the incident was escalated through the legal dispute that followed.

These events are distracting from the core business of providing excellent rail and supply chain solutions. By digitizing these assets, such events are quickly resolved, or better still, prevented entirely. Bottlenecks in processes are identified, and accountability is restored.

GPS is obviously a valuable component in the technology suite. It’s clear that GPS is not new, and it’s certainly no longer particularly costly. In the rail industry, all business stipulations are somehow linked to location. However, all GPS is not created equal. It must be integrated on the hardware side so it remains reliable in industrial environments, and it must be integrated on the software side to ensure the data it yields is cleaned with algorithms to ensure accuracy. It must be connected to other asset data and processes to ensure it yields knowledge for the user instead of information or “data soup” that creates further confusion and inaccuracies.

GPS is only currently working on locomotives and therefore “railcar compositions” when the loco is powered on. This is not all the time, so during the “power off” periods, the railcars and locomotive are “dark” and untraceable. This means that any collision, mishandling and malpractice are also hidden from view. Cutting-edge industrial IoT hardware with integrated energy management means that every single event is recorded and analyzed automatically for a period of six years. Precision and reliability enable automatic processes to be coupled to the software, opening the door to ideas like “dynamic contracts,” automated inspections and remotely managed checks and controls.

We can also switch our attention to operational performance in terms of using tools like geofences, dynamic ETAs, modal changes and fleet utilization, fleet distribution, fleet composition in asset type “mix” and other elements like monitoring partner services in switching, cleaning, maintenance times, idle times, “spot market” development, etc. The point is to extract more value from existing assets by improving efficiency and create service differentiation for competitive advantage at the same time. Surely, this is the holy grail of rail as we face the new challenges laid down by an increasingly complex and interconnected world.

In my next article, I will build further on extracting value from data and understanding how this technology can provide new ways of providing services to derive more value from the current fleet of railcars, equipment and components.

Daniel MacGregor is Co-founder of Nexxiot AG. He created and built this multi-million-dollar, digital supply-chain technology company from scratch. The company, he says “has a clear focus on digitizing mobile assets like rail freight and containers to create services for smoother operations in an integrated solution.” Daniel is a leading voice in the drive for sustainability and supports the creation of standards and applications. From hardware to information distribution and business-process innovation, Nexxiot’s clients deploy these solutions to differentiate their services and monetize digital insights. The company recently established a U.S. presence.

Tags: