Thanks to the advent of the latest innovations in artificial intelligence (AI) and machine learning (ML), smart cities — with a specific focus on the utilities sector — are undergoing unprecedented changes.
This article was originally published on Techopedia
The Capgemini Research Institute estimated that, together with the energy sector, the utility vertical can save between $237 billion to $813 billion USD from intelligent automation at scale.
Utility companies have been experimenting with AI use cases such as predictive maintenance, yield optimization, and demand/load forecasting. In 2019, more than half of energy and utilities organizations have deployed at least one practical implementation of AI technology, reaping its consistent benefits.
Even the public seems eager to enjoy the positive innovations brought forward by the AI transformation. According to a survey from PwC India, 71% of respondents were overwhelmingly optimistic about the chances that AI is going to help humans solve complex problems and live more enriched lives in the immediate future.
Let’s have a look at some actual use cases of AI and ML in the smart cities/utilities sector to evaluate the practical impact of these technologies from a pragmatic point of view. (Also read: Top 20 AI Use Cases: Artificial Intelligence in Healthcare.)
For public sector entities, such as large energy, power and utility companies, real-time information on energy usage can reduce wastage and loss, improve the efficiency of grid operation, optimize storage, and enhance predictive infrastructure maintenance.
Smart meters are an application of AI and ML which hold high potential in the energy and utilities field. AI, ML and the Internet of Things (IoT) form a crucial component of the government’s vision of smart cities and smart industrial zones. (Read more: 6 Tips for Securing an IoT Device.)
Cities provide a wealth of information that can be captured using IoT devices in real-time, including energy consumption. Power inputs can be adjusted automatically, leading to important savings, more secure supplies, and fewer outages.
Smart meters are useful even on a smaller scale as well. By using them, customers can tailor their energy requirements and thus reduce costs. The data generated in the process could be used for customized tariffs and more efficient supply.
In fact, in the United Kingdom alone, one electricity and gas smart meter is planned to be installed for every home and small business, for a total of 53 million by 2020.
Paris certainly is one of the most beautiful cities in the world — and has always been able to blend its marvelous historical and artistic tradition with modernity and innovation.
One of the most jaw-dropping ideas that the Paris administration has recently implemented, comes in the form of sensors placed on park benches to make them “smart.” The new IoT-enabled park benches are able to collect a constant stream of information that can be digested and analyzed by AI for a broad range of urban planning uses.
They can provide insights on environmental factors such as air quality by monitoring pollution levels, temperature, and atmospheric pressure. They can assist city planners by collecting data on frequency of traffic and space usage.
They can even “talk” with park habitués and tourists to ask for feedback on the facilities through a downloadable app.
The aging workforce issue is a problem which is going to have a profound impact in the natural gas and electric utilities. According to the US Department of Energy, 25% of U.S. employees will be ready to retire within five years, taking their skilled expertise away with them.
AI is able to tackle this issue by leveraging Natural Language Processing (NLP) and pattern recognition to process unstructured and structured data, and extract precious information from past internal communications, training documents, and technical notes.
IBM Maximo Equipment Maintenance Assistant leverage the power of IBM Watson AI to learn the “tribal knowledge” from experienced workers and transmit all this pooled information to the next generation of employees.
The intelligent machine can provide recommendations to equipment technicians and operators on how to perform complex maintenance tasks on the equipment. It can also capture feedback from technicians over time to finely tune its future recommendations, and help the whole workforce work more productively and safely.
Chatbots are everywhere. It’s not possible to talk about AI/ML use cases in any vertical without finding at least one “unique” chatbot among them. But in the case of smart cities, there’s a ton of them.
It doesn’t need a technician or a technology expert to see how chatbots are already making our lives easier by transforming how people communicate with brands. Consumers can already talk with brands on their own schedule, to buy things or get the answers they need on their preferred messaging service.
Cities are becoming “smarter” as nearly every store or brand is adopting a chatbot of some kind, yet some are more intelligent than others (or at least that’s what they say).
One of these examples is Maven, an AI engine developed by LivePerson that orchestrates conversational commerce interactions by blending together NLP, voice and text-based interfaces, and real human agents.
Computer vision to enhance security cameras is a pretty straightforward application of AI. Having a video where a crime is recorded is great, but unless someone knows something occurred, there is simply no reason to review the footage until that footage is eventually lost.
A surveillance system supported by a robust AI looking for patterns of criminal behavior is the equivalent to a team of detectives that never sleep analyzing all video in real time.
AI-enhanced security cameras can be used in schools and businesses to cut the response time whenever action needs to be taken. For example, if the person that needs to be detected is a “white male wearing a blue shirt,” the AI can differentiate between people entering an area who corresponds to the description and send an alert in real time.
His photos and video can also be uploaded directly to local first responders, who can find segments of videos which may contain him based on keywords (white, male, and blue shirt, in the above example) instead of having to search through hours of footage.
In Japan, an AI-powered security camera is so smart that it can estimate the poses of a suspicious person who is likely to commit a shoplifting crime.
In the parking space, AI has been used by companies such as Pixevia to integrate computer vision and advanced algorithms into a smart ecosystem.
By combining many advanced functions such as license plate recognition and pixel detection, off-the-shelf cameras can provide real-time information on space availability to customers and parking operators, and automatically enforce parking payment and duration.
Advanced algorithms can provide precise car position estimation, and predict parking usage during certain times of the day or night.
Eventually, this technology will be a perfect match for the upcoming autonomous vehicles which will be able to “speak” directly with parking lots and garages.
As Greta Thunberg (remember her?) keeps reminding us: the planet is facing imminent collapse if we don’t take appropriate measures to keep greenhouse gas (GHG) emissions under control.
AI may help us save the environment in many ways, especially by reducing energy wastes, land, water, and air pollution. It has been estimated that, by 2030, the application of AI technologies could reduce global GHG emissions by at least 4%.
In Singapore, IoT sensors and AI are being used to collect and analyze air quality and levels of pollutants and temperature in the city. This data combines with AI potentially to predict where air quality issues are and will be in order to ultimately mitigate the effects with effective preventative measures.
IBM researchers are testing a new form of AI to reduce the severity of air pollution in Beijing by analyzing data drawn from coal-fuelled factories, industrial complexes, weather conditions, and traffic congestion.
The system is able to provide useful insights on how to mitigate the effects of the Chinese city’s choking air pollution, such as temporarily closing certain factories.
Another place where AI can help with the reduction of GHG emissions is transportation systems.
Other than making autonomous cars possible in the near future, transportation can be made more sustainable even today. (Read more: Are These Autonomous Vehicles Ready for Our World?)
Researchers at the Department of Energy's Lawrence Berkeley National Laboratory are working on a computational tool based on deep reinforcement learning models called CIRCLES to smooth traffic in all congested cities.
CIRCLES — which stands for “Congestion Impact Reduction via CAV-in-the-loop Lagrangian Energy Smoothing” — simulates large amounts of vehicles driving in custom traffic scenarios.
This connected and autonomous vehicle (CAV)-enabled system can reduce energy consumption and improve traffic flow by reducing stop-and-go phantom traffic jams on freeways.
Used to make cities much more livable, advanced traffic control can also help reduce air pollution. Deep learning algorithms are used to combine satellite images with traffic information obtained from smartphones and environmental IoT sensors to improve air quality predictions.
AI-based automatic license plate reader (ALPR) software such as the one deployed by Rekor can also be used to provide recognize vehicles or for real-time detection of crimes and violations.
Probably one of the most amazing application of computer vision and AI is the creation of workplaces that are able to adapt to the needs of the employees to improve their wellness, especially in times like we're facing today with strict self-isolation measures being felt on a global scale.
Encompass AV has developed a system that uses AI to integrate HVAC, lighting, and security cameras together through data analysis. The system can make autonomous decisions to adapt and adjust the environment depending on activities registered or other external or internal factors.
For example: It can operate in circadian rhythm to make workers inside buildings feel as though they are outside, dimming lights to different color temperatures to simulate a particular time of day.
The system can also integrate with ceiling tiles that emulate a skylight in buildings lacking natural light. These holistic solutions can also identify needs and resources based on both real time and historical data to deliver autonomous responses that improve safety and increase energy efficiency.
Even developing nations understand the importance of employing cutting-edge AI for a leg up on energy development.
Beyond Limits is using a technology originated for NASA space missions to build the first cognitive plant in West Africa on behalf of Xcell, a Swiss-based global financial and minerals development company.
The efficiency of power generation from natural gas will be amplified many times-fold. The cognitive AI will be embedded in every part of the new power plant from the very beginning of its construction.
The efficiency and productivity of natural gas power plants have always been plagued by adverse environmental conditions, such as temperature and humidity. Beyond Limits’ AI is intelligent and aware enough to assist operators with its encoded expert-level human knowledge to make real-time adjustments.
Instead of producing unneeded power in excess that cannot be stored, the software can constantly monitor the gas turbines and the entire system to prevent wasted production or lose profit because of the underproduction.
Coupled with its ability to coordinate several interconnected production units at the same time, this futuristic power plant will be more efficient, productive, safe, and environmentally friendly.
One of the most frequently implemented applications of AI in nearly all industries and sectors, is the ability to make accurate forecasts and develop reliable prediction models.
Social and attitudinal factors are particularly salient in the case of new alternative energy products. Many energy and utility companies are using AI models to make these products accessible and interesting enough to consumers and governments, and predict their uptake.
In combination with IoT and big data, AI can analyze potential demand, typical pricing and government rebates which may affect the price, industry trends, as well as other social and geographical factors. (Read: What does big data do?)
Once all this data is digested, an accurate forecast can help companies estimate the investment costs and returns for the product, increasing the likelihood of a renewable energy sources to be accepted and implemented.
Providing robots with individual AI means facing several limits: processing power, space, and other flexibility constraints. CloudMinds found a creative solution to this problem by creating intelligence in the cloud — a place where humans and robots may combine their intelligence together and countless robotic “brains” can be hosted simultaneously. (Also read: 5 Defining Qualities of Robots.)
Their new cloud robot ecosystem is being built and enhanced by the collaboration of other AI service providers and developers to enhance intelligent capabilities, skills and robot applications.
The XR-1 Humanoid Robot is the first humanoid service robot powered by this new Cloud AI, and it’s learning at an amazing speed how to put its vision-controlled manipulation capabilities in practice.
At the current time of publication, it can already perform complex tasks such as threading a needle, open doors, shake hands, and even fold clothes.
Since its AI-brain is hosted in the cloud, the cute “android” can use much more powerful AI engines with low battery consumption and take full advantage of collective learning capabilities (a sort of “hive mind” shared by all robots), paving the way for a new generation of truly intelligent assistants for humans.
AI can be applied to strengthen public infrastructure by leveraging predictive maintenance to improve its availability, reduce accidents, and prioritize maintenance needs.
Pavement maintenance, in particular, is important to always keep roads safe, accessible, and secure, both inside and outside cities and towns. Maintenance involve different approaches, and choosing when it’s best to go for a routine rather than corrective or preventive intervention is never easy when budget constraints are always an issue.
The wrong decision may cost a lot more money in the long run, or expose vehicles to dangerous driving situations.
Most governments must prioritize their maintenance needs by performing visual or van-based inspections which are often dangerous, expensive, and highly subjective. The infrastructure technology company RoadBotics has implemented a revolutionary deep learning algorithm that uses computer vision to read road images and automatically make are objective and cost-efficient decisions on road conditions.
Their proprietary technology is saving a massive amount of taxpayer money by providing data-driven results to more than 150 governments across the world that can now fix the right roads at the right time.
ML brought massive changes in agriculture, and has already started revolutionizing this sector. Although the current AI use cases in agriculture are too many to be listed, it already brought a new level of scalability.
Smart flying drones are used to collect massive amounts of environmental data about fields and crops that is later fed to smart machines. No humans can process this amount of data and find the patterns that are instead clear to the AI. (Also read: Drones in 2020: What's Next?)
Machines are able to analyze data in real-time and make complex decisions on the spot, such as choosing or which fertilizer to use when to switch on irrigation, improving overall efficiency and sustainability.
They can also be used to run any number of simulations coming from proprietary or shared yields and predict the performance of crops.
We’re feeding data to the machines to help them feed us in exchange.
AI is growing up each day as a more integrated part of our cities’ future landscape. It is helping us shape a smarter and more efficient society, and has already proven to be one of the most helpful tools that computer technology brought to humanity.
With it, we can finally start addressing the larger issues faced by human society as a whole.
The Top 5 Use Case Categories for Edge Computing
How smart garbage bins powered by LoRaWAN can become an integral part of a smart city
Explainable AI: The Next Frontier in Human-Machine Harmony
Machine Learning Use Case: Facial Recognition Using Amazon Rekognition
Applying AI in Food Industry To Streamline Supply Chain Workflow
Please log in again. The login page will open in a new window. After logging in you can close it and return to this page.