Fincantieri, a global leader in shipbuilding and mechanical component design, wanted to shift from traditional reactive maintenance of turbogenerators to predictive maintenance with real-time remote monitoring. Using Microsoft Azure and Microsoft Azure Data Explorer (ADX), Fincantieri was able to do just that. Proactive monitoring enabled the company to postpone interventions by 25 percent and manage the storage of spare parts more effectively. The result is greater reliability and reduced downtimes for turbogenerators.
With over 230 years of history and more than 7,000 ships to its credit, Fincantieri is one of the world's largest shipbuilding groups and a global leader in cruise ship design. The Italian powerhouse is a leader in all high-tech shipbuilding industry sectors, from naval to offshore vessels, high-complexity ferries to mega yachts, and even systems and component equipment for mechanical and electrical segments.
Fincantieri's expertise extends worldwide through its 18 shipyards operating across four continents. One of its offerings includes a diverse range of steam turbogenerators, serving the versatile industrial market. As Francesco Malavasi, Business Development Engineer at Fincantieri, explains, "The industrial steam turbine market caters to diverse applications such as waste to energy and biomass power plants. Given the current fluctuations in gas prices and the continual evolution of the energy market, it's increasingly important to ensure the reliability of steam turbine and energy production.”
However, keeping these turbines running is challenging. The traditional reactive maintenance method often requires extensive travel, making it costly and inefficient. Any loss of operating time would also mean incurring prohibitive costs of supplemental energy, as the turbines often power the facilities themselves.
In a bid to adapt to the evolving energy landscape, Fincantieri embarked on a transformative journey in 2020. The company initiated a project to provide real-time predictive remote turbogenerator monitoring. “We wanted to expand our services to customers beyond standard maintenance activities. Hence, the new service includes not just data acquisition, but the application of algorithms which monitor the steam turbine 24 hours a day,” Malavasi sums up.
“By introducing real time customizable scripts, we enable Fincantieri to determine new rules or model new algorithms directly from the web,” explains Riccardo Zamana, Chief Technology Officer of BeanTech. “Together with Fincantieri, we’re commited to maintaining a high-performance managed service, consistently updated technological stack, and support for the adoption of artificial intelligence in the industry.”
Reinventing turbogenerator monitoring
Traditional turbine maintenance has limitations on data, which hinders monitoring. “The Programmable Logic Controller (PLC) installed in the power plant has limited memory for data storage. You can only access data from approximately the last six months, and older data is lost,” explains Malavasi. To address this, Fincantieri decided to use Microsoft Azure Data Explorer (ADX). The company’s headquarters connects to power plants via software installed directly at the plant, collecting the data in ADX for continuous monitoring.
ADX not only stores data but facilitates analysis, too. “When we access data directly from the power plant, we conduct a preliminary data analysis. This process previously took several hours or even days,” Malavasi explains. “With ADX, our algorithms perform a preliminary data analysis in a matter of minutes. This level of efficiency is not possible with a standard system.”
Aside from the PLC, Fincantieri collects data from other sources as well. On-site sensors—many of which Fincantieri developed for shipbuilding—help customers remotely monitor and interpret what is going on in the plant. With this, Fincantieri and its clients have massive amounts of data to sift through. This process is facilitated smoothly using ADX. Antonio Violante, IoT Project Manager at Fincantieri, explains, “ADX gives us the ability to easily find the relevant information with custom queries. The Kusto Query Language (KQL) is easy to learn and is extremely flexible—even our business users wanted to learn KQL to set up custom alarms.”
Turbocharging predictive maintenance
Fincantieri’s holistic understanding of power plants' operation enables the company to analyze any arising issues more thoroughly. “We can avoid two things. First, we can avoid sending people who are unequipped to solve the problem and reduce the need to send additional staff later. Second, customers don’t have to shut down the power plant while waiting for a technician to show up. This is crucial for them to prevent any economic loss,” explains Malavasi.
“With proper monitoring and predictions, we postpone major interventions by about 25 percent. This changes the power plant's business plan and cash flow, benefiting the customer.”
Francesco Malavasi, Business Development Engineer, Fincantieri
The solution also plays a pivotal role in managing the power plant's lifespan and critical spare parts inventory. “The lead time for receiving spare parts is very long. At the same time, any delays in repair and maintenance in a power plant can be disastrous. The power plants can choose to keep spare parts just in case, but keeping spare parts in stock is costly,” elaborates Malavasi. “It's not easy to account for all of that in your business plan.”
Fortunately, with ADX, Fincantieri can now help its customers manage warehouse inventory more efficiently. Continuous monitoring provides insights into which parts are wearing out and nearing their end-of-life, helping manage the ordering and stocking of expensive components. Historical data also helps plant operators avoid unnecessary repairs or replacements, potentially postponing major overhauls by two years. Malavasi shares, “We send the customer a plan in which all the supplies will be sent only when needed, based on monitoring and predictions. Because of this, we can postpone major interventions by about 25 percent. This changes the power plant's business plan and cash flow, benefiting the customer.”
All this has inevitably shifted the company’s relationship with their customers. “We’re not just contacting them when there's a problem. We reach out if something appears to be wrong even before an alarm is triggered,” Malavasi shares with pride.
Creating integrated digital solutions
For Malavasi, the continuous monitoring Fincantieri provides is one of the best in the market. “The main advantage of our system is that we can acquire the data from all over the system—the turbine, the boiler, the steam generator, everything,” he says. “Many predictive maintenance products only focus on one aspect, but our solution gives a bigger view.” Because Fincantieri not only monitors the turbogenerator but also the connected systems, it helps ensure maximum power availability and reliability for the 170 turbines it watches over. “With ADX's capabilities, we can monitor the performance of every unit effectively, optimizing our maintenance operations," Malavasi adds.
Fincantieri's predictive maintenance has received a positive reception, both internally and among its clients. And this is only the beginning. As Violante points out, “With ADX, we have a cloud platform where we collect data and monitor not only our steam turbines but potentially other assets in the future.” And with the ease of implementing their proprietary analytical algorithms with ADX, Fincantieri looks well-positioned to make the most of what cloud technology has to offer.
Follow Microsoft