The 16 Top Cross-Industry IoT Analytics Applications

Wojciech Prażuch

Aug 6, 2021 • 17 min read
IoT_Analytics_Applications

Every aspect of our lives is slowly transforming to be driven by next-generation technologies and IoT analytics is one of the leading ones.

Development of big data shifts us from conventional, rule-based systems to a futuristic world. How far are we from it?

Look at data science and AI-driven changes in the world. We unlock phones and perform transactions with face scans, create deep fake videos, and make our own emojis.

AI works as a translator (Google Translate to name one example among many), it can describe photos and videos, as well as create beautiful, artistic content. Machines have beaten the best player in Go, solved protein folding problems, and designed a website. All that in only eight years since AI and data science have started skyrocketing!

Even so, in many areas legacy systems have been doing just fine. But for how long? A data-centric approach to business flow indicates evolving along with the rest of the world and leaving the technological legacy behind.

IoT data analytics, a combination of Internet of Things, data science, and data engineering, supports this evolution in many industries. For cross-industry transformation, the sky's the limit with the technology we have. Or maybe not even that. In some cases, IoT analytics are already there.

Intelligent retail solutions

IoT data analytics are one of the change-drivers in retail that will probably result in traditional stores being outperformed by retail entities well adapted to emerging technologies. And it’s not a wave of the future, retail is an early-adopter of IoT and data analytics.

1. Queue detection

The answer to the most annoying part of in-store shopping. Thanks to queue detection, long queues will soon be a thing of the past. Providers such as Hikvision, Prolux, and Axxonsoft deliver these solutions based on Convolutional Neural Networks.

CNNs are specialized neural networks that enable automatic measurement of shopping queues’ length. Aligned with IoT analytics, they provide real-time data and latest insights about the number of customers queuing to make a purchase or try chosen products and instantly analyze it to optimize in-store traffic management.

2. Video analytics

Queue detection is based on real-time video analytics. Its general idea is processing video feed to detect objects or people and process it further.

It can be used to analyze shop traffic, like counting the number of customers in-store or measuring time spent inside. IoT data analytics also help track how customers act and what products they choose.

This data provides valuable and actionable insights into customers’ behavior and improves decision-making processes for retail managers on how to handle their clientele.

On top of that, video analytics act as a product heatmap to analyze the placement of products on shelves and identify product shortages.

3. Just-walk-out shopping

The “just-walk-out” idea, a breakthrough approach to shopping, was originated by the Amazon Go supermarkets and is set upon self-service stores without registers. Customer identification is based on a palm scan connected to a customer's card. Basically, they enter, get the needed products, and the purchase is executed as soon as they leave the store.

The trick is in video analytics, again. “Just-walk-out” shops use it to track and analyze customers’ shopping carts and what products are in them.

IoT analytics in agriculture

Farming, the oldest industry, has a crucial need for efficiency, especially considering the growing global population. Climate change has made it harder than ever to forecast crop yields. From unpredictable weather conditions to higher cost of resources, it’s not an easy job to do these days.

Although the knowledge of how to adjust cultivation methods to environmental challenges has increased greatly, it has not been widely adopted. We have the know-how, but it often goes unused.

IoT analytics may change this state of affairs by giving the agriculture industry new monitoring tools.

4. Smart water management

Water consumption and management are a constant struggle for farmers. They juggle between keeping crops watered and overpaying for water, which is already scarce.

Instead of acquiring additional water sources, we can optimize distribution. If we provide crops with the exact amount of water needed, exactly when and where the watering has the biggest efficiency, we can conserve the supply.

IoT analytics can introduce data-driven measurement that saves thousands of tons of water by increasing watering efficiency. Solutions monitoring soil moisture data can automatically optimize the environment for the crops. This way we can ensure maximum growth with lower costs of watering and water use.

5. Forecasting fertilizer usage

Although fertilizers are a boon to big scale farming, their low cost means that they are often overused. Simply put, farmers often use too much of them when less would be enough.

However, the negative side effects of fertilizers (such as contribution to climate change) may increase their costs. IoT devices and data analytics can help to reduce these expenditures and optimize fertilizer usage simultaneously.

By monitoring the condition of soil and crops, farmers can reduce the amounts of fertilizer used, cut costs, and ensure optimal conditions for the crops. There are already sensors available for that.

6. Cultivation recommendation systems

Temperature and humidity sensors are able to generate data for the annual picture of the climate in a chosen location. This data can feed data science systems and machine learning algorithms to recommend the most effective crops for farming in terms of yields and profits.

Such a system could use IoT analytics to gather and analyze data from farm sensors. Cleaned data would feed machine learning algorithms, which in turn would create predictive analytics and forecasts for the crops.

Business intelligence could add value to these systems by creating visualizations that would enable farmers to check the current state of the market and respond to its changing needs.

This data-driven approach might seem like a revolution in the agriculture market but could turn out to be beneficial for both farmers and consumers. Larger crops would drive food prices down and still make the farms as profitable as possible.

Smart cities

Smart cities have countless applications for IoT analytics, and yet more is to be discovered! Some of the biggest and best-to-live cities have already adopted IoT analytics and data science as a way to improve living conditions.

The flow of this data is the same. It is transferred to a data lake to be cleaned, preprocessed, and transformed using data engineering technologies. Then, it goes to a data warehouse, a hub that is connected with all the smart applications like predictive analytics that retrieve data to create machine learning algorithms.

For easier deciphering of the processed data, it is presented with business intelligence tools like dashboards to be more readable and digestible.

Custom applications can analyze and present nearly every aspect of public life. From traffic jam forecasts, through road maintenance notifications, to preventing utility failures, smart city applications are a huge area of opportunity.

7. Traffic and energy management

Who wouldn’t like the opportunity to avoid traffic during rush hours. Singapore, with its density of population reaching 8000 people per square kilometer, supports better time management among its citizens. The city collects operational data about traffic to reduce time spent in jams.

Another example, Amsterdam, will launch driverless taxis and buses at the end of 2023. The data from such a vehicle will be collected by IoT device and sent to a data lake for further analysis to identify potential improvements in their method of operation.

8. Energy management

Smart cities have huge power demands and one of the biggest challenges for IoT analytics here is to make sure the energy flow for all citizens is stable and reliable.

Oslo involved IoT connected devices in car charging stations to plan deployment of new charging units that ensure best energy utilization among the citizens.

Copenhagen, on the other hand, created a complex solution for traffic, air quality, waste management, and energy use to optimize resource distribution.

Singapore also collects energy consumption data in certain districts to optimize energy production and ensure high energy capacity in the city.

9. Smart solutions for public life

Singapore and Oslo also use smart lighting. Street lamps adapt light intensity to current lighting during the day, which helps to reduce costs for city utilities.

Copenhagen connects its parking systems with street lights, charging stations, and buildings to create a fully data-driven city. This centralized system allows the city to plug in new IoT devices and connect them within the central data lake. The system also notifies cyclists what speed they should ride to come across the green light on the next crossing.

Dubai has come even further by creating autonomous police stations, where citizens can pay fines or report incidents. Police officers can focus on preserving peace thanks to automating the bureaucracy. What’s more, the data collected at the autonomous stations allows further data mining for the city.

There are also places where citizens can report graffiti and potholes in the city using IoT device data and their analytics, Boston among them.

Many companies offer smart city solutions. For example, T-mobile offers out-of-the-box waste management that is already deployed in many cities across the world.

Data-driven healthcare

Healthcare is one of the industries that can benefit the most from the IoT advancements. There are many IoT applications that could improve overall effectiveness in patient treatment, as well as save costs and time, if IoT analytics were globally adopted. What are their capabilities?

10. Health condition monitoring

All medical devices, like those for monitoring blood sugar, blood pressure, oxygen levels, heart rate, ECG, or weight, if connected to healthcare facilities, would ensure immediate treatment if a patient’s health deteriorated. This could help patients to avoid hospitalization or operational treatment and reduce costs of treatment by detecting danger in early stages.

Another case is monitoring and treating sleep dysfunctions during hospital stays. Some diseases could be automatically detected using IoT sensor data and machine learning, as ML proved successful in many medical applications.

Some health monitoring devices are already deployed in the form of smart bands or smartphone apps. They monitor sport activity or sleep quality. The data generated by these IoT applications can support people in staying healthy. Although this case isn’t linked directly to medical facilities, it also serves as an IoT application and could be used as a regular health insight in the future.

11. Intelligent healthcare facilities management

Healthcare IoT can also be responsible for effective hospital and medical facilities management. IoT analytics applications are able to inform hospital management about bed capacity, patients’ condition and their documentation, medical device availability, or streamline operations such as appointments. This would help greatly in managing medical equipment and improving operational flow.

Hospitals could also use a data warehouse to store the IoT health monitoring data of home-based patients. The data would be next transformed and cleaned in data engineering pipelines, then reworked into business intelligence dashboards to find patients that need medical aid. Complemented with current workload and equipment/staff occupation data, it’d provide the managers with an overall image of their facility’s status.

Data from all hospital departments and laboratories could optimize how patient appointments are scheduled. In a common data lake, they could be preprocessed and unified to create a single, consistent database of hospital management.

To optimize appointments even more, an IoT device could inform about queues or even enable remote clinical services. If a meeting is delayed, IoT analytics could monitor the queue length to inform the patient in real-time to save his time and reduce the number of waiting patients.

12. Smart data acquisition for medical research

Another aspect to be handled by IoT analytics is research, as medical data is heavily fragmented. Healthcare-oriented IoT could finally unify the data and accelerate medical research.

The biggest obstacle in performing medical research is sourcing data. Each medical facility has only its own data, whereas data originated elsewhere, if accessible, could provide new insights. Digitizing data acquisition could create common data lakes for medical researchers and speed up the development of new treatment techniques, vaccines, or drugs.

Industry 4.0

Production lines depend on full availability of the machines. If one fails, the whole line can be compromised and production delayed, leading to lost money. Therefore, monitoring and controlling production machinery is crucial for ensuring high capacity. This is definitely something that IoT analytics can help with.

13. Industrial machinery control and monitoring

Keeping industrial machinery in good health is essential for production flow consistency. By using IoT analytics, industrial companies can ensure the highest level of malfunctioning detection in early stages and avoid serious crashes.

Each machine connected to the data lake sends its data there, hence is monitored. If one of them starts malfunctioning, the incoming failure is detected by IoT analytics quickly and repaired instantly.

This approach prevents major breakdowns and production delays and enables companies to avoid money losses caused by these complications and the costs of repairing broken machinery.

14. Data-driven production line optimization

Moreover, the whole production process can be improved and optimized with the gathered data so the overall throughput and effectiveness would increase.

Siemens created a platform dedicated to publishing industry IoT applications called MindSphere. The platform is in use on many production lines across the globe and provides out-of-the box BI for IoT devices. The whole production process is monitored in the cloud and the dashboards can be available to all users.

Whether it’s a power plant, oil refinery, a car production facility, or any other production-oriented facility, MindSphere can be adopted there. It integrates very well with Siemens machines located on-site.

Coca Cola has adopted MindSphere in their factory in Austria. Thanks to IoT system for sensor data, the downtime of machines has been significantly reduced, and the anomalies were quickly detected and handled. The whole investment was returned after only 6 months.

IoT analytics for real estate

Although not very obvious, real estate is quite a large business when it comes to IoT analytics. The possibilities are countless - from energy reporting, through appliance control, to security.

15. Controlling smart devices

The biggest players in technology are already developing smart devices controlled with smartphones. Smart light bulbs, radios, TVs, but also smart locks, which can be remotely opened in applications. Latch, Zerv, or ButterflyMX can serve as a few examples.

When it comes to surveillance, cameras can be easily controlled and checked with mobile applications. Moreover, using machine learning, a surveillance camera becomes smart and detects movement or dangerous events recorded in the feed.

16. Increasing effectiveness of smart buildings

What is more, there is wide adoption of IoT analytics for smart buildings. For instance, energy production in photovoltaic panels can be monitored day-by-day, thus giving information about savings, possible upgrades, or detected malfunctions of the infrastructure.

Humidity and temperature sensors located across a house can give information used to adjust AC or heating. The temperature in each room can be adjusted separately. The total cost of heating can even be evaluated on a per-room basis.

Unfortunately, the whole technology is currently heavily fragmented and there is no common interface for storing all the data in one place. Hopefully, one day we will be able to monitor and control everything using a single application.

The future is here

Every domain is evolving day-by-day and IoT analytics play a key role in that transformation. As a technology and consulting company, we are happy to support companies in this process with our expertise in both IoT and data science. Yesterday’s ideas are tomorrow’s solutions and this couldn’t be more true for IoT and data science!

Related topics

More posts by this author

Wojciech Prażuch

State of Machine Learning in Fintech Report