22, Feb 2024
Dark Data Steps Into the Infrastructure Spotlight
Sensors are everywhere – turning off highway lights when roads are empty, monitoring the health of bridges, and monitoring the intricate dance of telecommunications networks and electrical grids. Every flicker of these sensors is a byte of data, meticulously logged and stored. With the costs of data storage plummeting over the past decade, we’re talking about an avalanche worth of data digitally warehoused.
Much of this data has been resting in the dark, unanalyzed and unseen. This is what experts call dark data. And now, as AI steps into the infrastructure arena, this dormant data is about to step into the spotlight.
“Indeed, there appears to be an enormous amount of data collected on infrastructure operations that could be better used to improve their effectiveness,” said IEEE Life Senior Member Raul Colcher.
AI thrives on data – the more, the better. And when it comes to training sophisticated AI models, this dark data, collected over years from myriad sensors and systems, may be extremely valuable.
So, what’s the big deal about bringing this dark data to light? For starters, it’s a game-changer for infrastructure operations. With AI algorithms churning through mountains of previously unused data, we can expect leaps in efficiency and new ways to design and use our infrastructure for a future where data moves more frequently than people.
BYTES BUILD BETTER
Much of the time, dark data isn’t used because it isn’t properly tagged, and is therefore difficult to analyze. Some research suggests that the machine learning algorithms that allocate resources within mobile phone networks could be greatly improved with the use of dark data. In another case, data scientists at an oil and gas plant were able to use dark data to improve a digital model of the plant without disrupting operations.
WHERE WILL THE IMPACT BE FELT THE MOST?
The benefits of analyzing and modeling this data are vast and varied. From planning to operations, maintenance, and beyond, every facet of infrastructure could see a transformation. Picture more accurate models, better automation, and a deeper understanding of how our systems truly work.
CHALLENGES ON THE HORIZON
But, it’s not all smooth sailing. Dark data, while abundant, isn’t always clean or error-free. Questions of bias, data provenance, and security loom large. How we address these challenges will be crucial in unlocking the full potential of AI in infrastructure.
“The surge in data quantity doesn’t guarantee better results,” said IEEE Member Qi Qi Wang. “Filtering out disruptive or poor-quality data presents a substantial challenge.”
Learn more: 2023 was a landmark year in AI, as broad swaths of the public became more aware of AI thanks to the power of generative AI tools. IEEE Spectrum covered developments in-depth. Check out their run-down of the top AI stories in 2023.
- 0
- By Rabindra
14, Feb 2024
Driven by Data: Fueling EVs
Imagine a world where the hum of engines and the smell of exhaust are things of the past. That’s the future electric vehicles (EVs) promise us. As it stands, our roads contribute about 12% of global carbon emissions, but the rise of EVs is key to turning this around and making transportation cleaner and greener.
But, to accelerate EV adoption, the sector needs to leverage a new type of fuel altogether: data.
EVs are game changers for infrastructure around the globe. With their growing numbers, they’re reshaping everything from city layouts to how we manage our electricity supply. The shift to EVs brings new challenges for our power grids and the need for innovative infrastructure to keep them running.
Here’s how data can help meet those challenges:
THE GRID
The central challenge that data can solve boils down like this: Lots of people tend to charge their EVs at the same time, placing sizable strain on electricity grids. But there are also periods when few people are charging their cars, meaning there is excess capacity in the grid. Is there a way to shift EV owners’ behaviors so they charge during times of lower demand?
“Intelligent EV charging is becoming a big area,” said IEEE Senior Member Kyri Baker. “Charging EVs at high power rates can strain infrastructure like transformers, so smart scheduling of charging can help extend the lifespan of these components.”
By looking at how customers have used power in the past, it’s possible to make the distribution of electricity – especially for charging things – more efficient. By figuring out the busiest times and places for electricity use, the people running the power grid can spread out the power usage better. This way, they can avoid overloading the system, save on energy costs, and make the whole network run smoother.
WHERE CAN YOU CHARGE?
Knowing how much battery power an EV has left is super useful, not just to individual drivers, but to all their fellow drivers on the road. That information, sometimes referred to as the vehicle’s state of charge, can tell an individual motorist how far they can go before they need to refuel. By collecting and analyzing state-of-charge data for many vehicles, a driver can know the best charging station to use to avoid a wait. And builders would also be able to use the data to understand the best place to build new charging stations.
“By analyzing the historical data of electric vehicle charging station use, like time of day, day of the week, seasonal variations, etc., it is possible to understand where the demand is high,” said IEEE Senior Member Marcio Andrey Teixeira. “The behavior of the data is another important factor because it provides insights like preferred charging times and popular routes. This information helps in the optimization of the placement of charging stations along frequently traveled routes.”
Learn more: Electric vehicles are here to stay, according to an editorial from IEEE Power & Energy Magazine. This means that the distribution grid and its stakeholders need to ensure that EVs and the grid work together. The Nov.-Dec. 2023 issue devotes numerous articles to the challenges of integrating EVs.
27, Oct 2023
Artificial Intelligence in Its Many Forms Will Be the Most Important Area of Technology in 2024
India, October 27, 2023 — IEEE, the world’s largest technical professional organization dedicated to advancing technology for humanity, today released the results of “The Impact of Technology in 2024 and Beyond an IEEE Global Study,” a new survey of global technology leaders from the U.S., U.K., China, India, and Brazil. The study, which included 350 chief technology officers, chief information officers, and IT directors, covers the most important technologies in 2024 and future technology trends.
Telecommunications, Manufacturing, and Financial Industries Most Impacted by Technology in 2024
The top five industry sectors that will be most impacted by technology in 2024, according to survey responses, are:
● (41%) telecommunications (as compared to 40% in 2023)
● (39%) manufacturing (as compared to 30% in 2023)
● (39%) banking and financial services (as compared to 33% in 2023)
● (31%) automotive and transportation (as compared to 39% in 2023)
● (31%) energy (as compared to 33% in 2023)
AI will be the Most Important Technology in 2024 – Used in Diverse Ways,
Across the Global Economy
What areas of technology will be the most important in 2024? From over a dozen areas of technology, when asked to select the top three, respondents chose:
● (65%) Artificial Intelligence (AI), including predictive and generative AI, machine learning (ML) and natural language processing (NLP)
● (28%) Extended reality (XR), including metaverse, augmented reality (AR), virtual reality (VR) and mixed reality (MR)
● (24%) cloud computing
Other important technologies in 2024 include 5G (22%), and electric vehicles (20%).
In 2024, AI applications and algorithms that can optimize data, perform complex tasks, and make decisions with human-like accuracy will be used in diverse ways, the study finds. Of top potential applications for AI next year, technology leaders surveyed selected:
● (54%) real-time cybersecurity vulnerability identification and attack prevention;
● (42%) Increasing supply chain and warehouse automation efficiencies
● (38%) aiding and accelerating software development, automating customer service
● (35%) automating customer service
● (34%) speeding up candidate screening, recruiting, and hiring time
● (32%) Accelerating disease mapping and drug discovery
● (31%) Automating and stabilizing utility power sources
Survey participants were asked what percentage of jobs across the global economy in 2024 will be augmented by AI-driven software, and 26-50% of jobs was cited by 41% of those surveyed. Over one-quarter (28%) cited 1-25% of jobs; another 26% cited 51-75% of jobs, and 5% cited 76-100% of jobs.
Benefits of Extended Reality (XR), Digital Twin Technologies, 5G and 6G
According to the IEEE survey, virtual simulations using extended reality (XR) and digital twin technologies to more efficiently design, develop, and safely test product prototypes and manufacturing processes will be important in 2024 (63% very important, 29% somewhat important).
Respondents see 5G benefitting the following areas the most in 2024, including greater benefits to transportation infrastructure and sustainability as compared to 2023.
● (54% in both 2024 and 2023) telemedicine, including remote surgery, health record transmissions
● (46% in 2024 vs. 49% in 2023) personal and professional day-to-day communications
● (46% in 2024 vs. 56% in 2023) Remote learning and education
● (43% in 2024 vs. 51% in 2023) entertainment, sports and live event streaming
● (39% in 2024 vs. 29% in 2023) Transportation and traffic control
● (27% in 2024 vs 25% in 2023) manufacturing/assembly
● (30% in 2024 vs 23% in 2023) carbon footprint reduction and energy efficiency
Close to nine out of 10 of global technologists (88%) agree 6G will primarily be an evolving work in progress in 2024, but will be standardized in the next 3-5 years. In addition, a strong majority (94%) of global technologists agree that the development of communication satellites for mobile connectivity will bring parity to some rural and developed regions globally in 2024.
Don’t Count Quantum Out
Generative AI may continue to dominate the technology landscape, but other technologies such as quantum will have significant, if less-noticed impacts (87% agree, including 51% who strongly agree). Furthermore, 86% of respondents agree, that in 2024 quantum computing will gain the most attention for significantly higher computing power – a trillion times higher than that of today’s most advanced supercomputers, as well as for its application to post-quantum cryptography and cybersecurity.
Technologies that foster sustainability continue to be a priority. A strong majority agree (92% agree, including 54% who strongly agree) that they have prioritized sustainability goals for 2024 and beyond, and any technologies their company implements are required to be energy-efficient and help shrink their carbon footprint.
Cybersecurity Concerns Continue
The cybersecurity concerns most likely to be in technology leaders’ top three in 2024 remain the same as last year: data center vulnerability (43% in both 2024, and 2023), cloud vulnerability (42% in 2024, down from 51% in 2023) and security issues related to the mobile and hybrid workforce/employees using their own devices (41% in 2024, down from 46% in 2023). However, the study finds other cybersecurity concerns will rise in 2024, including ransomware attacks (37% in 2024 up from 30% in 2023), phishing attacks (35% in 2024 up from 25% in 2023), and insider threats (26% in 2024, up from 19% in 2023).
