In early May, Israel’s Iron Dome missile defense system intercepted 35% of the 690 rockets and mortar shells launched by Hamas from the Gaza Strip. The system was able to prioritize threats in real-time so effectively that only three civilians lost their lives.
The Iron Dome uses a "system of systems" approach, which Col. Natan Barak of the Israel Defense Force helped develop. In 2003 he retired to commercialize the approach to managing real-time threats for industry use by founding mPrest, which provides mission critical monitoring and control software for Industrial Internet of Things applications. Manufacturers may not face such literal life-and death decisions as Israel does on a constant basis, but they similarly rely on thousands of sensors and devices to make tough choices in an instant. We emailed Barak, also the CEO mPrest, to find out how this technology help shield industry from the daily barrage of problems.
NED: Can you describe how you helped develop the algorithms now used for Iron Dome in the early 2000s and what the inspiration was?
Natan Barak: During my military service, I led the development of a generic platform that offered visibility across the full operations of battle ships, submarines, combat planes, helicopters, naval commando, and other supporting systems. This platform integrated and shared data into a holistic, real-time view for decision makers, offering maximal flexibility and logic that adjusts itself to the conditions in real time. The Iron Dome Command and Control system has similar, yet much more complex tasks–it needs to take into account various and ever-changing threats, multiple data sources in real time, the location of our forces and planes, weather, and other conditions of the terrain, all in real time to deliver immediate decisions and results.
Beyond that, the ability of the platform to handle distributed and adaptive image building algorithms, along with real-time interaction plans, are what makes this robust, successful solution of Iron Dome, which we developed together with Rafael.
NED: How did you turn that into mPrest?
NB: The idea behind the mPrest platform is built on the same foundation as the Iron Dome command and control system. We wanted to create a platform that could seamlessly integrate with existing systems, gather extensive, relevant data from these systems, and use AI-driven analytics to make decisions in real time. We call it a ‘System of Systems,’ or orchestration and optimization platform. We provide end-to-end optimization, instead of only a limited, local solution.
Most, if not all, enterprises have multiple siloed systems that do not communicate with each other. mPrest empowers its customers to leverage and optimize by connecting all systems and resources. We believe that if you are not integrated, you are not optimized. Our ability to identify patterns and anomalies enables us to offer enterprises optimization similar to what we created in the defense domain.
In our projects with Netafim, for example, the capability of our ‘System of Systems’ is demonstrated perfectly. Netafim has integrated sensors on their irrigation pipes and now has the ability to analyze the data gathered from the pipes, transforming them into a big-data company rather than an irrigation company. Our platform gathers data from multiple sources, including soil quality, irrigation status, nutrients, weather forecast, etc., and makes ongoing decisions and recommendations for farmers based on pre-defined rules and definitions, such as when to water the field and when it is time to harvest.
NED: So what is being tracked and intercepted in the factory?
NB: There are processes that are not yet fully optimized in any enterprise and production environment, but new IIoT connectivity and monitoring capabilities are being leveraged to offer increased efficiency, reduce energy costs, and enable predictive maintenance to avoid malfunction and breakage.
Essentially, we receive all inputs, make a decision of whether or not an object represents a danger, select the best interception scenario, and then manage its execution. In most cases, there are multiple enemy rockets launched. Combining all relevant systems, analyzing the data, and making a decision in real time becomes a big data analytics algorithm.
Another way of describing our system is as a real time, distributed asset management and analytics platform. Once described in these terms, the connection to industrial applications, and specifically to energy management applications, becomes quite evident.
Over recent years, due to decentralization and de-carbonization, there is an exponential growth in distributed clean energy resources (photo voltaic cells, batteries, electric vehicles, etc.). Current systems are not designed to cope with the amount of resources, with energy flow being multi-directional, or with the negative effect on system quality, in parallel to the good they bring. In other words, energy utilities require a new product, which is capable of aggregating data from all the various sources, forecasting needs (supply and demand) and managing these distributed assets in real time.
NED: Concerning Industry 4.0, what are the key benefits of using mPrest’s smart monitoring, analytics and control applications, and what are the advantages of using this solution over more established versions?
NB: We believe that if you are not integrated, you are not optimized. Our ‘System of Systems’ approach allows for full end-to-end integration of all other systems (including the Advanced Distribution Management Systems). This is a capability that other analytics companies, including the larger vendors, do not offer. The ADMS itself is an integrated system, but mPrest is an overall integration of all systems, including those outside of the ADMS.
All of mPrest’s solutions are built on an open, micro-services-based architecture that enables swift integration with existing systems, rather than requiring the replacement of legacy systems altogether, such as in the case of the larger vendors. This significantly reduces both risk and cost. And the platform offers inherent scalability and the ability to add new services in the future without effecting the architecture. This allows for fast time to market, which shortens the ROI. mPrest is also the only proven cloud implementation, which is also important from a cyber security perspective.
For the energy industry, for example, we are the only vendor that offers both an advanced platform that optimizes grid management, with end-to-end visibility and control, in addition to having what we believe to be the most advanced application on the market. The mDERMS application solves the specific challenges of utilities in the age of the energy revolution - enabling utilities to effectively manage the plethora of Distributed Energy Resources (DERs) and their relations with the grid, microgrid integration, and asset health management.
NED: Do you have any specific examples of the system preventing catastrophic failure, and conversely, in day-to-day operation, how much power is being saved (or not wasted)?
NB: mPrest’s Asset Health Management and specifically our Transformer Health Management are operational systems, monitoring power transformers 24/7. As such, the very fact that power transformers are being managed via our application is reducing critical failure rate significantly for companies like New York Power Authority (NYPA). Our project with NYPA also won EPRI’s Technology Transfer Award. Of course, many other transmission utilities have power transformer failures. Implementing mPrest’s product would bring such critical failures to a halt and save tens of millions of dollars per year per such utility.
A critical failure of a power transformer entails many cost considerations, including: a new power transformer, which typically takes between 12 and 24 months to replace. During that period, there is a need to source energy from alternate sources. As this is an unplanned energy procurement, costs are significantly higher than normal energy procurement. This can drive a typical total cost of a power transformer to be anywhere between $10 million to $15 million.
In addition to the above, such a critical failure can result in deaths, service level interruptions, and negative publicity, which utilities would want to avoid.
NED: As more devices become connected, how can that increase the value of systems like mPrest’s?
The magnitude and significance of the optimization will grow over time, as millions of DERs connect to the grid, and their impact on the operational systems grows in significance. On the other hand, if utilities do not implement DERMS in parallel to integrating all their IT and OT operations, they are basically flying blind–they will fall behind in terms of reduced profit, customer satisfaction, and increased operational costs.