AI BI and IoT

Business is enabling the BI, AI & IoT to help the production lines to purchasing trends to consumer analysis. I am Ravindra Pande (ravindrapande@gmail.com  /  rrpande@indiatrainingservices.in) trying to extrapolate the possibilities in scientific way this is not Si-Fi but current situation and very near future situations. Like to have your suggestions and feedback as always. Imagine a smart future! A future where machines are not merely dumb devices but intelligent creations that can work in tandem with human beings. A future that looks remarkably like the robotic utopia in I, Robot (Well, except the homicidal robots!). This future is not merely an imagination but a natural consequence of the two most dynamic technologies of today.

Factories deploy AI to automate complex physical tasks requiring adaptability and agility. Marketers use AI to generate individualized recommendations to the automatic order fulfillment scenarios. The list is expanding really fast and open to all new ideas and error free efficient process. A host of services taken for granted today, from credit card fraud detection to email spam filters to predictive traffic alerts to personalized reminders, wouldn’t be possible without AI. One area where AI is used extensively is business intelligence. Enterprises leverage deep learning algorithms to spot behavioral patterns likely to lead to sales, use cues from IoT sensors for predictive maintenance and inventory optimization and do more. However, what businesses do now is just the tip of the iceberg of possibilities.

AI Enables Live Decision Making, with the proliferation of data, several businesses run the risk of data overload. The unprecedented growth of Big Data and the obsession to analyze such data can easily gag the core operations of the business. AI-powered business intelligence software enables enterprises to break down data into manageable insights, and make sense of Big Data.

AI also has the potential to change the dynamics of analytics. Conventional data analytics focused on descriptive analytics or analyzing data to report what happened. The present generation of AI-enabled analytics tools enable predictive analytics or using data to decipher future insights. This, however, is based on “best guesses” with behavioral and historical data used to guess probabilities.

Prescriptive analytics is all set to take over in the near future. AI-powered prescriptive analytics tools would scour through vast swathes of data and enable users to prescribe various possible actions and advise viable solutions. Prescriptive analytics not just predicts, but offers sound advice as well, and explains why things will happen the way it will or does.

The shift from reactive predictive analytics to proactive prescriptive analytics improves the potency and relevance of business decisions. Live, real-time insights enable enterprises to make the best use of their operational data, making decisions based on what’s currently happening rather than based on what happened in the past. Much of the recommendations can be automated as well, with the best course of action determined by the intelligent machine based on the available inputs.

AI Brings Voice and Facial Recognition to the Center stage : AI-powered voice-activated digital personal assistants have already enamored millennial in a big way.  The spurt in deep learning-powered applications such as speech recognition interfaces, its widespread adoption by businesses and the tremendous popularity of digital voice assistants such as Apple Siri, Amazon Alexa and Google Assistant are portents of things to come. Voice will replace the keyboard and touch interfaces as the default norm for individuals to engage with brands, cutting across industries.

Likewise, matured facial recognition technology is all set to make big strides from present levels, in the near future. AI-powered facial recognition technology may just make the highly irritating password obsolete.

AI Powers Hyper-Personalization : AI-based intelligence learns from experience, becoming better with each experience or transaction. With the next prescribed decision automatically better than the previous one, the stage where the AI model is highly matured and covers all eventualities isn’t far off.

It gets better. AI-powered systems of the future could automatically decipher the user and even the users’ emotions from the soon-to-be-commonplace voice commands, to make highly accurate recommendations or engage with them at a truly personal level. The next wave of AI-powered assistants will be capable of analyzing huge troves of data contextually, in real-time, to grasp customers’ need and priorities quickly, and do what’s required. AI is all set to make hyper-personalization the default norm, rather than a premium service as it is now.

At a macro level, enterprises would be able to collate information from various data points and make real-time live sentiment analysis. For example, an enterprise could collect live data from the customer’s engagement with the company, their social media posts and other data, to understand their thought process and emotional reaction about a product and make real-time interventions to either reinforce or change such perceptions.

AI is already helping industries such as financial services, healthcare, securities trading and life sciences in a big way. For instance, AI is taking over the role of the clinical assistant, helping physicians make faster and more reliable diagnoses. Such instances will become commonplace to the extent human intervention will become rare.

However, as of now, machines traditionally don’t do well when it comes to abstract tasks involving human capabilities such as empathy, creativity, judgment, inspiration and leadership. Two critical management functions, innovation and managing people, are still almost entirely with humans. This could change in the future thanks to AI systems becoming more mature. Presently, algorithms may suffer from some amount of bias or subjectivity, considering the algorithms are designed by humans after all. As training data gets more mature, such biases and negative effects will be quickly eradicated.

Artificial intelligence is here to stay. AI has the potential to transform how top executives make decisions, how marketers engage with customers, how enterprises compete with each other and how they develop overall to become more potent and powerful. The future of business intelligence will surely be driven by AI-enabled systems.

Well, Artificial Intelligence deals with the creation of systems that can learn to emulate human tasks using their prior experience and without any manual intervention. (Basically Intelligent Systems!). Internet of Things, on the other hand, is a network of various devices that are connected over the internet and they can collect and exchange data with each other enabling the AI in an exponential scale.

To understand this further, all these IoT devices generate a lot of data that needs to be collected and mined for actionable results. This is where Artificial Intelligence comes into the picture. Internet of Things is used to collect and handle the huge amount of data that is required by the Artificial Intelligence algorithms. In turn, these algorithms convert the data into useful actionable results that can be implemented by the IoT devices.

Without AI-powered analytics, IoT devices and the data they produce throughout the network would have limited value. Similarly, AI systems would struggle to be relevant in business settings without the IoT-generated data pouring in. However, the powerful combination of AI and IoT can transform industries and help them make more intelligent decisions from the explosive growth of data every day. IoT is like the body, and AI the brains, which together can create new value propositions, business models, revenue streams and services

The Power for IoT : The number of sensors in the Internet of Things (IoT) will grow like never seen before we’re talking of billions to trillions of sensors. As most of them will be connected wirelessly, we need to rethink the technologies for how to power the sensors.

Batteries are part of our mobile lives. We find them in each portable device and in millions of remote controls for smart home and building automation. With trillions of wireless IoT sensors, the demand for batteries will increase into infinity. However, infinity is no option for a battery-powered IoT.

Limits of Lithium: For the 10 trillion wireless sensors delivering the needed data for IoT, they would require one million tons of lithium – the combined worldwide lithium production in 10 years. And, we even need much more for our smartphones, electrical cars, local energy storage systems, etc. Consequently, there’s not enough lithium in the world for all of these applications.

Additionally, the environmental impact of lithium mining results in water shortage, air pollution and destruction of nature reserves. No recycling technology exists today that is capable of producing enough pure lithium for a second use in batteries.*

Lithium is just one element batteries contain. There are also toxic heavy metals such as mercury, lead, cadmium and nickel in batteries, which are detrimental to the environment. At the end of their lifetime, they need to be disposed of carefully and expensively. Recycling isn’t always an option and is polluting as well. Reports reveal that it takes 6 to 10 times more energy to reclaim metals from some recycled batteries than from mining.

Challenges of Changing : Besides the environmental impact of battery production, disposal and recycling, there are further costs that we need to consider as batteries need maintenance.

Wireless connectivity supports devices to be flexibly installed or mobile and their location needs to be documented and updated as their location changes. In a large building system, hundreds of sensors are distributed over several floors and offices. Often, the devices are mounted unobtrusively in places that are difficult to reach, e.g. on or above drop ceilings. Depending on the battery technology in use, a user will dispose of between 200 and 1,600 batteries over 20 years in a residential home with only 50 nodes.

In addition, each device has a different battery access method and requires different types of batteries. This results in extra work, making the battery replacement a challenging and time-consuming effort. Usually, batteries don’t run out of energy at the same time. So, the technician might just have left the facility after changing some batteries when the next battery dies. Well, call the technician again.

Benefits of Battery less: Considering the costs of batteries, IoT, with its trillions of sensor nodes, needs a more eco-balanced and maintenance-free alternative to power mobile devices. This alternative exists today and is already deployed in building automation and smart home systems or in outdoor environmental monitoring systems: energy harvesting wireless sensors.

Like we switch our power production to the use of renewable energy sources, such as sun, wind, water, etc., self-powered sensors use the same principles of harvesting energy from the surrounding environment at a micro-level. Miniaturized energy converters use kinetic motion, light or temperature differences to power wireless sensing and data communication. There’s no need for a battery change and disposal and no need for time-consuming maintenance. Simply install and forget.

It’s a simple calculation and our ecological responsibility to realize a self-powered Internet of Things. Energy harvesting wireless sensors are the only way to avoid tons of battery waste and to ensure reliable and maintenance-free system functionality.  Artificial Intelligence and the Internet of Things is like a match made in Tech Heaven.

While both of these disciplines have individual value, their true potential can only be realized together. There are many different applications across multiple industries that require Artificial Intelligence and Internet of Things.

Connected Robots: Ever wanted the help of a robot? Well, that’s exactly what you will get with Collaborative Robots or Cobots. These Cobots are highly complex machines that are designed to help humans in a shared workspace with environments ranging from office to industrial. They can be a robot arm designed to perform tasks or even a complex robot designed to fulfill tough tasks without least manual interventions.

All favorite Drones: Drones are aircraft without a human pilot (The piloting is done by the software!). They are extremely useful as they can navigate unknown surroundings (even those beyond the reach of the internet) and reach areas hazardous for humans such as offshore operations, mines, war zones or burning buildings.

Smart Cities: When everything is getting smart, why not whole cities? Smart cities can be created with a network of sensors that are attached to the physical city infrastructure. These sensors can be used to monitor the city for various civic factors such as energy efficiency, air pollution, water use, noise pollution, traffic conditions, etc. In India specifically the great innovative vision of Hon Prime Minister Mr. Nerandra Modi enabled a great project with in Indian country to enable smart administration of cities.

Smart Retailing: This is shopping made smart! AI and IoT can be used by retailers to understand the customer behavior (by studying the consumer online profile, in-store inventory, etc.) and then send real-time personalized offers while the customer is in the store. While Artificial Intelligence in the Internet of Things is a relatively new concept, it has already been successfully applied in many real-world applications. (Yeah, this world is more tech-savvy than we thought!) Some of these applications are given as follows:

Self Driving Cars: Sound like futuristic science-fiction yet they are very much a part of today’s reality. The Tesla Motors self-driving cars use the latest advancements in Artificial Intelligence and the Internet of Things. While these cars are still in the testing phase (With multiple legal and ethical concerns as baggage!) they are still one of the easier innovations of IoT. A unique feature of the Tesla self-driving cars is that all of them act like a connected network. Whenever one car learns some new information, which is passed on to all the other cars. And that is used to predict the behavior of cars and pedestrians on the road in various circumstances.

Endangered Species Preservation: There are many animals that are endangered or going extinct in various countries (No thanks to human of course!). Also, the traditional methods of tracking these animals with collars are stressful and dangerous (Both to the animals and researchers). So WildTrack’s footprint identification technique (FIT) uses IoT and AI algorithms to identify the species, individual, age and gender of an animal from its unique footprint. Then this data can be used to recognize patterns relating to animal movements, species population, etc. that help in preserving various endangered species.

Smart thermostat : Everything is becoming smart these days, this device uses IoT to allow temperature checking and controls from anywhere using smartphone integration. It is also quite simple to use, which is one of the primary reasons for its success (apart from AI and IoT of course!).  Artificial Intelligence plays a big role in the Nest Labs thermostat. It is used to understand the temperature preferences of the users and also their daily schedule. Then it adapts accordingly for optimal temperature and also maximum energy savings.

Automated vacuum cleaner – iRobot Roomba: When everything else is becoming smart, why not a smart vacuum cleaner? The iRobot Roomba is developed by three members of MIT’s Artificial Intelligence Lab and it uses IoT and AI to clean a room as efficiently as possible. It is a robotic vacuum cleaner that uses a set of sensors to detect obstacles, dirty spots on the floor or even steep drops such as stairs.

So, it essentially remembers the layout of the living space (As much as machines can anyway!) and then uses the most efficient and economical movements for cleaning. A smartphone app can be used to adjust the performance requirements with “Clean” mode, “Spot” mode, “Dock” mode, etc.

Conclusion: This is an exciting new time to live in (both for humans and machines!). With multiple advances in artificial intelligence, light-speed communications, and analytics, IoT is even more convenient and high-performance IoT devices are taking over almost every domain of technology. Moreover, the declining hardware costs make it feasible to embed sensors and connectivity in just about any device imaginable. Taken together, Artificial Intelligence and Internet of Things are ushering in a new era where “smart” is just the normal state of being and the robotic utopia in the future appears more and more attainable in the present.