The use of AI and machine learning is another example of how the company continues to innovate its aftermarket solutions to help customers move to 100% planned maintenance environments – avoiding unplanned events and reducing costs.
“Through AI we are further enhancing our capabilities to see patterns and preventive maintenance insights in the massive data set our oil analysis technology provides,” said Irene Makris, vice president, Customer Service, Pratt & Whitney Canada. “With this deeper understanding of what’s going on inside the oil-wetted components of the engine, we are able create a ‘know early, respond early’ environment for the customer that helps manage risk, and prevents disruptions in service.”
This kind of preventative maintenance allows customers to avoid unscheduled engine removals and other maintenance events by providing visibility on potential issues as much as a thousand hours in advance. The oil analysis technology is available on a majority of Pratt & Whitney Canada engine models including the PW210 and has analyzed over fifty-five thousand oil samples from customer engines.
Based on the guidance provided by the oil analysis technology, dozens of events across the fleet have transformed into scheduled interventions, dispatch availability has increased, and the solution has saved PW210 customers over US$1 million in averted secondary damage.“Our oil analysis technology was launched in 2017 and enables proactive and predictive maintenance by identifying potential issues before they occur without intrusive inspection,” said Nicolas Chabée, vice president, Marketing, Helicopters, Pratt & Whitney Canada, “AI makes the task of comparing oil samples with past data simpler, more efficient and easily scalable.
”Oil analysis technology is part of Pratt & Whitney Canada’s suite of digital engine services and is another proof point to the company’s one-stop shop approach to simpler, more cost-effective maintenance management, designed to maximize aircraft availability, upkeep and planning, and reduce operating costs.
More information about Pratt & Whitney Canada’s oil analysis solution for helicopter engines including which engine models are supported is available here.
WANT MORE INSIDER NEWS? SUBSCRIBE TO OUR DIGITAL MAGAZINE NOW!
3D printing experts have given us their forecasts for the additive manufacturing trends to watch and also the future of 3D printing.
Of course, 3D printing does not exist in a vacuum (except when it does). Figuratively speaking, additive manufacturing must be considered in relation to the bigger picture. Whether that is the immediate 3D printing ecosystem of materials, machines, and software or the wider world of technology, economic and social forces.
Therefore, the final question we asked the industry leaders was what non-AM/3DP frontier technology do you see as the most significant for the coming decade and why?
As noted in earlier articles, there is a tendency to bristle against marketing terminology. However consider how earlier terms such as the Internet or the Cloud proved useful in crystallizing a vision and differentiating such technologies from the earlier physical, and digital, elements they were formed from.
Extending this to today’s buzzwords, will Decentralized Autonomous Organizations, Web3, the Metaverse, or Non-Fungible Tokens advance beyond their initial manifestations?
Alternatively, and to coin a phrase, are some of these technologies all extruder and no filament?
The responses reveal an industry optimistic about the future and the role of technology within it. The second most mentioned theme was a concern for the climate and how technology can be harnessed to address net-zero goals for carbon. Within this theme, the experts discuss renewable energy such as hydrogen, energy storage including solid-state batteries and recycling, and electric vehicles.
Also featuring prominently was the use of advanced visualization and collaborative tools, specifically Augmented Reality (AR) and Virtual Reality (VR). How (and if) the Metaverse comes to be manifested for the mainstream user remains to be seen. Will Zuckerberg’s Facebook mutate into Neal Stephenson’s Global Multimedia Protocol Group, or will the currently cumbersome nature continue to deter wider adoption?
Data and the software needed to generate useful information or applications are considered by the experts, whether in manufacturing execution systems, the digital thread, the digital marketplace, or big data. Receiving fewer mentions were distributed ledger technologies and FinTech – here NFT’s, decentralized networks and blockchain-based systems were on the expert’s radar – but of note to fewer respondees. Likewise, while fascinating topics, synthetic biology, and materials received only a couple of mentions.
But the most frequently cited technology, mentioned by a quarter of our experts, was Artificial Intelligence (AI) and Machine Learning (ML).
McKinsey’s most recent State of AI report shows service operations, product and service development, and marketing and sales as the top three areas where AI is currently applied. Our respondents look towards enhanced generative design tools and cloud-based intelligent systems underpinning the future of advanced manufacturing.
We’re seeing machine learning/artificial intelligence playing a more important role in additive manufacturing – powering software that is underlying the entire manufacturing workflow. Not only is machine learning optimizing organizations’ use of 3D printing, but it is also optimizing other advanced manufacturing technologies, including robotic welding, machining, finishing, and inspection operations, in full production environments. While this technology is playing an important role within our industry, I believe it will play a significant role more broadly within society over the coming decade.
We’re already seeing how cloud-based intelligent systems are making our lives easier, and more efficient. If we’re ordering take-out through an app, or a ride via one of the well-known rideshare services, machine learning/AI is already playing a role. We’re also seeing this technology’s influence in healthcare, whereby virtual visits with a physician are removing the need to take time out of our busy schedules to physically drive to a medical office. Instead, we can request a virtual appointment, and through the power of our laptop or smart device, from the comfort of our own homes, both patient and physician can discuss and assess symptoms for the physician to render a diagnosis and prescribe a treatment. Over the coming decade, I think we’ll see the role of machine learning continue on its upward trajectory, influencing in a greater way how we interact with one another and closing the physical gap to bring society virtually closer.
Beyond additive manufacturing and 3D printing, mixed reality (MR) will fast become a significant technology within the industry, and over the next decade will become the norm for customer service support. With the exponential rise in digital manufacturing fueling the industry, customers have less time for service calls and higher production runs to meet, therefore the way in which companies deliver services in this ever-evolving business and work environment needs to change fast. MR can do just that, providing real-time support that can guide PSPs through troubleshooting as they work, ensuring self-sufficiency and smoother operations around the clock.
In this context, HP has partnered with Microsoft to launch the first-ever industry mixed reality service – HP xRServices. The collaboration will see HP xRServices and Microsoft HoloLens 2 create a virtual-real world combination in which customers can connect with HP engineers in a split second through mixed reality, advising them on any issue, at any point of their print production. Wearing the Microsoft HoloLens 2 headset and supported by HP xRServices solution, users will get the feeling of being physically present with a virtual coach on hand to guide them through the process, meaning no time wasted on long service calls, resolutions are swift and press downtime is kept minimal.
WANT MORE INSIDER NEWS? SUBSCRIBE TO OUR DIGITAL MAGAZINE NOW!
By using a combination of real-time data analysis in the cloud and machine learning at the edge, industrial companies are steadily turning their aspirations into realities and spurring the next industrial revolution.
Google Cloud and Siemens has announced a new cooperation to optimise factory processes and improve productivity on the shop floor. Siemens intends to integrate Google Cloud’s leading data cloud and artificial intelligence/machine learning (AI/ML) technologies with its factory automation solutions to help manufacturers innovate for the future.
Data drives today’s industrial processes, but many manufacturers continue to use legacy software and multiple systems to analyse plant information, which is resource-intensive and requires frequent manual updates to ensure accuracy. In addition, while AI projects have been deployed by many companies in “islands” across the plant floor, manufacturers have struggled to implement AI at scale across their global operations.
By combining Google Cloud’s data cloud and AI/MLmachi capabilities with Siemens’ Digital Industries Factory Automation portfolio, manufacturers will be able to harmonise their factory data, run cloud-based AI/ML models on top of that data, and deploy algorithms at the network edge. This enables applications such as visual inspection of products or predicting the wear-and-tear of machines on the assembly line.
Deploying AI to the shop floor and integrating it into automation and the network is a complex task, requiring highly specialised expertise and innovative products such as Siemens Industrial Edge. The goal of the cooperation between Google Cloud and Siemens is to make the deployment of AI in connection with the Industrial Edge—and its management at scale— easier, empowering employees as they work on the plant floor, automating mundane tasks, and improving overall quality.
“The potential for artificial intelligence to radically transform the plant floor is far from being exhausted. Many manufacturers are still stuck in AI ‘pilot projects’ today – we want to change that,” said Axel Lorenz, VP of Control at Factory Automation of Siemens Digital Industries. “Combining AI/ML technology from Google Cloud with Siemens’ solutions for Industrial Edge and industrial operation will be a game changer for the manufacturing industry.”
Check these articles out:
Industry 4.0 Needs Horizontal Integration
Industrial Robots VS Cobots—Which Is Right For You?
Siemens Improves 3D Printing And Scanning Workflows
For other exclusive articles, visit www.equipment-news.com.
WANT MORE INSIDER NEWS? SUBSCRIBE TO OUR DIGITAL MAGAZINE NOW!
FOLLOW US ON: LinkedIn, Facebook, Twitter
Having a predictive maintenance plan in place, powered by machine learning, will give you unprecedented insight into your operation and will lead to serious benefits in efficiency, safety, optimisation, and decision making. Article by Richard Irwin, Bentley Systems.
One of the goals of reliability is to identify and manage the risks around assets that could fail, causing unnecessary and expensive downtime. We know it is important to identify areas of potential failures and rate them in terms of likelihood and consequence. We have put good reliability strategies in place and have implemented proactive condition-based maintenance programs. Today, machine learning is helping maintenance organisations get to an elevated level of situational intelligence to guide actions and provide early warnings of impending asset failure that previously remained undetected. Machine learning is paving the way for smarter and faster ways to make data-driven decisions in predictive maintenance.
While machine learning has been researched for decades, its use in applying artificial intelligence (AI) in industrial plants and infrastructure asset operations is now advancing at a rapid pace. This influx of using machine learning is due to the growth in big data, the Industrial Internet of Things (IIoT), computing power, and the need for superior predictive and prescriptive capabilities required to manage today’s complex assets. While machine learning has typically been linked with industries such as transportation and banking (think self-driving cars and fraud monitoring, respectively), there are many uses for machine learning and predictive maintenance within the industrial sector. This article will focus on some of the principles within machine learning and industries that are primed to take advantage of the application of machine learning to maximise the benefits it brings to improve situational intelligence, performance, and reliability.
Before starting, it is important to point out that there are many options and techniques available to gain more insight and make better decisions on the performance of your assets and operation. It all comes down to knowing what the best fit is for your needs and what type of data you are using. Data comes in many shapes and sises and can consist of time-series, labelled, random, intermittent, unstructured, and many more. All data holds information, it’s just a case of using the right approach to unlock it, and this is where the algorithms used within machine learning help decision makers.
It is important to understand the complexity involved with machine learning before you make a decision on what is appropriate for you and your organisation. Here are some questions to ask yourself before implementing machine learning:
Machine learning makes complex processes and data easier to comprehend, and it is ideal for industries that are asset and data rich. In any industry, the ability to recognise equipment failure, and avoid unplanned downtime, and repair costs, among others, is critical to success. This is even more relevant in today’s turbulent times. With machine learning, there are numerous opportunities to improve the situation with predictive maintenance and the ability to predict critical failures ahead of time.
Predictive maintenance will be one of the most applicable areas where machine learning can be applied within the industrial sector. Predictive maintenance is the failure inspection strategy that uses data and models to predict when an asset or piece of equipment will fail so that proactive corrective actions can be planned in time. Predictive maintenance can cover a large area of topics, from failure prediction, failure diagnosis, to recommending mitigation or maintenance actions after failure. The best maintenance is advanced forms of proactive condition-based maintenance. With the combination of machine learning and maintenance applications leveraging Industrial IoT (IIoT) data, the range of positive outcomes and reductions in costs, downtime, and risk are worth the investment.
Whatever path is chosen, the benefits machine learning can offer to big data are only just being brought to fruition. Opportunity is rapidly developing with productivity advancements at the heart of the data rich industry in which you work. While healthcare, financial, automotive, oil and gas, electric and power, and water utility sectors are already advancing with machine learning, there’s another sector leading the way in this fast-moving digital transformation: manufacturing.
Manufacturing has always been the main industry when mentioned alongside machine learning, and for good reason, as the benefits are very real. These benefits include reductions in operating costs, improved reliability, and increased productivity—three goals that relate to the holy trinity of manufacturing. To achieve this, manufacturing also requires a digital platform to capture, store, and analyse data generated by control systems and sensors on equipment connected via the IoT.
Preventative maintenance is key in improving uptime and productivity, so greater predictive accuracy of equipment failure is essential with increased demand. Furthermore, by knowing what is about to fail ahead of time, spare parts and inventory can use the data to ensure they align with the prediction. Improving production processes through a robust condition monitoring system can give unprecedented insight into overall equipment effectiveness by monitoring air and oil pressures and temperatures regularly and consistently.
This example is centred around a steel manufacturer who routinely shuts down operations to perform maintenance on its assets, which is very costly. The steel output can sometimes warp or “crimp” during the production process as it travels through different stages. These failures can only be corrected every six months (as well as monthly for smaller fixes) during planned—and very expensive—maintenance that involves long periods of downtime. The main goals of applying machine learning here were to: reduce defects and locate root cause; identify key variables that matter the most; and prioritise assets during shutdown.
The first part of the machine learning process was to sort the data into a self-organising map using neural networks to organise data into 10 distinct classes based on parameters of the steel, such as thickness and weight, as they entered each manufacturing stage. Other techniques included decision trees to learn the pattern of data and to identify which features were important in those patterns; asset health prioritisation to provide ranking; asset health indexing to determine the health of the assets; principle component analysis to reduce the dimensionality of the data; and clustering/anomaly detection, which highlights how each stand deviates from its normal operating mode.
What developed was a method for dealing with different types of products, the ability to identify the top variables associated with production defects, and a process for applying anomaly detection to equipment in an industrial plant. It was shown that these processes could reduce the need for extensive analysis of equipment and give operators better tools and more insights to make maintenance decisions. A significant amount of time is spent locating the cause of the issues and performing maintenance. The new algorithm can be run before planning the shutdown, and it can identify which stand to prioritise during shutdowns through analysis of the asset anomaly charts. Focusing on assets that are the most at risk optimises the shutdown, as it is only conducted for a limited time.
Early adopters of machine learning are already reaping the benefits of predictive maintenance in the speed of information delivery, costs, and usefulness. This gives you more information and insight to make smarter decisions. Bentley Systems’ users are combining machine learning with Bentley’s other digitalisation technologies to make this process even more beneficial—by making it model-centric and adding visualisation dashboards, cloud-based IoT data, analytics, and reality modelling to machine learning, the result is a complete solution for operations, maintenance, and engineering. Machine learning can also be leveraged within digital twins to provide even more predictive insights.
Having a predictive maintenance plan in place, powered by machine learning, will give you unprecedented insight into your operation and will lead to serious benefits in efficiency, safety, optimisation, and decision making. The digital transformation for industry is now at a tipping point, with technologies all converging at the same time—a predictive maintenance approach to reliability and asset performance means that root cause analysis (RCA) could be a thing of the past. Machine learning takes into consideration the whole history of failures and identifies the signs of failure in advance.
Read more:
Growth Of The Digital Twins Market Is Driven By Industrial Digitalisation
Collaborative Robot Market To Exceed US$11 Billion By 2030
WANT MORE INSIDER NEWS? SUBSCRIBE TO OUR DIGITAL MAGAZINE NOW!
FOLLOW US ON: LinkedIn, Facebook, Twitter
According to a report by Tractica, titled “Artificial Intelligence Market Forecast”, the global artificial intelligence (AI) software market revenue is expected to increase from US$9.5 billion in 2018 to US$118.6 billion by 2025. The study includes market sizing, segmentation, and forecasts for 315 AI use cases distributed across 30 industries. The steady growth of the AI market in the consumer, enterprise, government and defence sectors can be observed as applications of AI technologies and solutions are becoming a reality.
“While the market is still a few years away from an inflection point for real growth, it is critical for both end users and solutions providers to identify the technologies and use cases where they want to invest in AI,” commented Aditya Kaul, research director at Tractica.
AI use cases covered by this report includes three main categories: vision, language and analytics. Vision and language represent the perceptive brain which aims to enhance speech and sight capabilities. While analytics represent the analytical brain which deals with extracting and processing raw data, using traditional machine learning techniques for example. Although analytics and big data are huge drivers of the AI market, pure analytics only represent 35 percent of revenue from AI use cases. In fact, the main driver of the market is actually language and vision use cases in combination with analytics, representing 65 percent of the revenue. New AI use cases in the manufacturing sector includes supply chain optimisation, human-robot collaboration, digital twins and robotic and machine vision enhancements.
FOLLOW US ON: LinkedIn, Facebook, Twitter
READ MORE IN OUR LATEST ISSUE!
WANT MORE INSIDER NEWS? SUBSCRIBE TO OUR DIGITAL MAGAZINE NOW!
Nanyang Technological University, Singapore (NTU Singapore) and AMD have joined forces to launch a Data Science and Artificial Intelligence Lab, which will nurture next-generation tech leaders with the latest industry-driven digital skills.
The S$4 million joint Data Science and AI Lab will leverage AMD’s deep-learning technologies and NTU’s global strengths in machine learning, artificial intelligence (AI) and data science, to complement the university’s Data Science and Artificial Intelligence undergraduate programme. NTU students will be exposed to real world applications such as developing software algorithms used in security fields like identification and motion detection. They will also work on big data and analytics which are now frequently used by leading organisations. For example, students will get opportunities to develop clinical support solutions using big data analysis to aid medical diagnosis. Students will also undergo training to participate in supercomputing competitions using AMD’s versatile open source software, Radeon Open Compute (ROCm) platform. This software facilitates ultrascale or hyperscale computing, a form of high-performance computing that can simulate complex systems within just a few days where it typically used to take years.
Potential research projects and real-world machine learning and deep learning applications that students are expected to work with in the lab will leverage AMD’s open software platform for accelerated computing. For example, with the ROCm open software platform, Tensorflow (open source machine learning framework) users will benefit from graphics processing unit (GPU) acceleration and a more robust open source machine learning ecosystem. The collaboration will also provide NTU students with local and overseas attachment opportunities with AMD. From the next academic year in 2019, NTU undergraduates will be able to intern at AMD’s Shanghai Research and Development Center (SRDC) and the Singapore Product Development Center (SPDC). NTU will also support AMD’s engineers who wish to pursue PhD programmes through EDB’s Industrial Post-graduate Programme (IPP). “The philosophy behind AMD’s open source deep-learning architecture is about flexibility and choice – for developers and academics to have a range of hardware vendor-agnostic tools to innovate at scale. NTU is the premier institution for artificial intelligence, and we’re very excited to bring our technologies to empower the participating students to freely explore the diversity of deep-learning applications,” said Allen Lee, Corporate Vice President for Asia Datacenter Group, General Manager for China R&D Center, AMD. NTU’s Dean for the College of Engineering, Professor Louis Phee, said, “This collaboration with AMD highlights NTU’s drive in nurturing strong relations with top industry partners to provide students with industry-relevant education. This will not only give them first-hand insights in solving real-world challenges, but also give them an edge when navigating today’s dynamic workplace environment.”
AMD will provide one instance of AMD’s current generation of server processor (EPYC) and Radeon Instinct MI25 accelerator that can be scaled to handle hyperscale workloads to start. AMD plans to provide the university with its latest server technologies, such as the Radeon Instinct MI60 accelerators in 2019 and beyond. AMD’s AI and machine learning experts will work closely with NTU professors to conduct joint training and workshop sessions for industry members.
The collaboration will also support Singapore’s digital initiatives such as the AI.SG programme and support the nation’s transition into industry 4.0.
“The establishment of the AI and Data Science Lab between AMD and NTU today is a great example of how industry and academia can leverage transformative technologies like AI to develop real-world solutions. This partnership is testament to the company’s confidence in Singapore and the innovation cluster we have here. Such an effort will also deepen Singapore’s talent pool for AI, and provide Singaporeans with the necessary skillset to participate in the opportunities that rapid technological adoption is bringing,” said Mr. Ling Yuan Chun, Deputy Director, Semiconductors, Singapore Economic Development Board.
WANT MORE INSIDER NEWS? SUBSCRIBE TO OUR DIGITAL MAGAZINE NOW!
