Taxonomy of IoT Usecases: Seeing IoT Forest from the Trees

IoT comes in many forms. Variation of use cases seems endless. IoT devices itself has many types and can be arranged in different configurations.

Following are of those device classes.

  • Ad-hoc/ Home/ Consumer (Embeddables , Wearables, Holdables, Surroundables, see Four types of Internet of Things?)
  • Smart Systems – ( they monitor the outside world, have lot of small sensors, have hubs that connect via Zigbee or cellular and connection from hubs to cloud)
  • M2M/ Industrial Internet (Sensor and inbuilt, often pre-designed)
  • Drones and Cameras (Never underestimate the most ubiquitous IoT device, video Cameras)

Those devices can be used to solve a wide range of problems. Obviously, it is hard to do a complete taxonomy, yet writing even a subset down would help us lot with understanding IoT.

The taxonomy is arranged around people, and each level moves further away from individual and becomes high level. Different levels are categorized from personal (e.g. wearables) to macro-level control ( smart cities). The following picture shows each category.

IoTUsecaseTaxonomy

Let us look at each category in detail.

1. Wearables

Wearables are devices that are with you. They range from pills you might swallow, a Fitbit, a watch, to your mobile phone. The goal of these use cases is to make your life better.

  • Health: Fitbit, personal health (e.g. Incentives for good habits)
  • From asset tracking to smart signage, and safety
  • Sports – digital coach, better sport analytics
  • Facial Recognition with real-life analytics and interactions

2. Smart Homes

These use cases try to monitor and improve your home giving you peace of mind, comfort, and efficiency.

  • Energy efficiency, smart lighting, smart metering, smart elements, smart heating, smart rooms, bedrooms
  • Integration with Calendar and other data, deriving context, and take decisions and drive the home environment based on current context.
  • Safety and security via home surveillance, monitor health and kids, perimeter checks for pets and kids etc.
  • Smart gardens (e.g. watering, status monitoring)

You can find more information from 9 Ways A Smart Home Can Improve Your Life.

3. Appliances

Appliances have a duel role. On one hand, they provide new experiences to the end user, hence play a role in Smart Home. On the other hand, they provide better visibility and control of appliance to the manufacturer. Devices include your car, smart lawn mowers, kettles etc.  Most products will have a digital twin, that will provide analytics and important information both to the consumer and the manufacturer.

Following are some use cases.

  • Products can interact with users better, optimize, learn and adapt to the user (e.g. smart washers and dryers that notify when done and product displays been replaced with apps)
  • Better after sales services, better diagnosis, remote diagnosis ( efficient customer support), faster update and critical patches
  • Adaptive and proactive maintenance as needed. With IoT, products can monitor themselves and act if there is a problem
  • Using product usage data to improve product design.
  • Get some appliances ( e.g expensive ones like load mower) under a pay per use model rather than buying them.
  • Know the customer better: better segmentation, avoid churn ( if he is not using it, find out)
  • Hobbyists/ Entertainment (e.g. drone racing, drone cameras)
  • Advertisements via your appliance (e.g. refrigerator let you order missing food via a App, and the manufacturer may charge for recommendations they made from companies)

HBR article, How Smart, Connected Products Are Transforming Companies, provide a good discussion about some of the use cases.

4. Smart Spaces

Smart spaces use cases monitor and manage a space such as a farm, a shop, forest etc. It would involve pre-designed sensors as well as ad-hoc sensors like drones etc. Often camera’s computer vision also plays a key role.

Following are some of the use cases.

  • Smart Agriculture (watering based on moisture levels, pest control, livestock management), correlate with other data sources like weather and delivery of pesticides etc though drones.
  • Surveillance ( wildlife, endangered species, forest cover, forest fire)
  • Smart Retail: Smart stores ( sensors to monitor, what gets attention), fast checkouts (e.g. via RFID), customer analytics for stores, In store targeted offers via smartphones, better customers service at the store.
  • Quick service restaurants(QSR) – measure staff performance & services, improve floor plan & remove bottlenecks, optimize queue & turnover
  • Smart Buildings ( Power, Security, Proactive Maintenance, HVAC etc)

For related use cases, see How The Internet of Things Will Shake Up Retail In 2015 and The Future Of Agriculture?

5. Smart Services Industries/ Logistics

These use cases use IoT to improve the services industry and logistics. They focus on monitoring and improving underline processes of those businesses. Following are few examples.

  • Smart logistics and Supply Chain( tracking, RFID tags)
  • Service industries: Airlines, Hospitality etc. The goal is efficient operations, and visibility (e.g. where my baggage?) and proactive maintenance.
  • Financial services, Smart Banking, Usage-based Insurance, Better data for Insurance, and Fraud detection via better data
  • Better delivery of products via Drones
  • Aviation – Report, find the problem, and find the fix, parts before plane lands,
  • Telecommunications networks

6. Smart Health

Smart health will be a combination of wearables, smart home, and smart services. This would include use cases like better health data through wearables, better care at hospitals, in-home care, smart pill bottles etc that would monitor and make sure medications are taken, and better integration of health records.

7. Industrial Internet

The Idea of the industrial internet is to use sensors and automation to better understand and manage complex processes. Unlike smart spaces, these use cases give owners much for flexibility and control. Most these environments already have sensors and actuators installed. Most of these use cases predate IoT and falls under M2M.

Following are some use cases.

  • Smart manufacturing
  • Power and renewable energy (e.g.Wind Turbines, Oil and Gas)  operations and predictive maintenance. The goal is to add value on top of existing assets (takes about 40 years to replace) .
  • Mining
  • Transport : Trains, Busses
  • HVAC and industrial machines

You can find more use cases from GE’s making world 1% better initiative.

8. Smart Cities

Smart Cities ( and my be Nations) brings everything together and provides a macro view of everything. They focus on improving public infrastructure and services that make the urban living better.

Following are some of the use cases.

  • Waste management, smart parking ( e.g. find parking spots)
  • Traffic management ( sensors, Drones), air quality and water quality, smart road tax
  • Security: Surveillance, gunfire sensors, Smart Street lightings, Flooding alerts,
  • Smart buildings (energy, elevators, lighting, HVAC), Smart bridges/ constructions(put lot of sensors into concrete etc)
  • Urban planning

You can find more information from articles How Big Data And The Internet Of Things Create Smarter Cities, and Smart Cities — A $1.5 Trillion Market Opportunity.

Conclusion

As we saw, use cases come in many forms and shapes and likely they will get integrated with and change our lives at many different levels. This is the reason that analysts have forecasted an unprecedented number of devices (e.g. 15-50B by 2020) as well as a market size (e.g. 1-7 Trillion by 2020) for IoT that dwarfed any earlier trends like SOA or Big data.

Following are few observations about the use cases.

  • Each use case tries to solve a real problem. They do this by finding a problem, instrumenting data around it, and analyzing that data and providing actionable insights or carrying out actions.
  • Some use cases are enabled by creative sensors, such as using camera to measure your heart rate or sensors mixed into the concrete while building a bridge.
  • Analytics are present in almost all use cases. One of the key, yet often unspoken assumption is that all data get collected and analyzed later. We call this batch analytics.
  • However, lot of use cases need realtime decisions and sometimes need to act on those decisions. There have been many efforts on relatime analytics, but comparatively less work has been done regarding acting on the decisions.
  • These use cases might lead to other use cases such as showing related advertisements on your appliance or on the associated mobile App.

Hope this was useful. I would love to hear about if you thoughts about different categories and use cases.

Advertisements

What do we need from an IoT Analytics Platform?

Internet_of_Things
IoT is considered to be the main driver behind analytics for next few years. Predictions point to billions of devices given rise to new innovative use cases.
What do we need in an IoT analytics platform?  Of course, you can take any analytics platform and build on top of it, which happens right now. However, IoT analytics has some special features such as time series nature, geo locations, devices etc., that let us build more out of the box behaviour.
Here is what I think should be done.
Level 1: Single Device Level Analytics.
This is showing the data about a single device in graph, map, and sending an alarm if it is too high.
  1. When a user selects a device in the UI, show default set of analytics such as showing its location in the map, show how each attribute (e.g. temperature) behaved against time.
  2. Integrate a Chart (e.g. Bar Chart) generation wizard so that the users can build their own charts and add them against the device. These charts should update automatically as new data arrives at the system.
  3. Let the user click on a chart and setup alerts from the charts or UI. Those alerts need to deployed in a realtime analytics engine like CEP.
Level 2: Analytics across multiple devices 
Let user aggregate data across all, groups, and at different levels of the hierarchy
  1. Have default aggregated views for location and known attributes. For example, show all devices in the group in a map or show all device values for one attribute in a line chart against time.
  2. Add support for groups and aggregation in the chart generation wizard integration as well when a user is building his own charts.
  3. Let users write their custom queries to aggregate data (e.g. using SQL-like language such as SparkSQL) and deploy them in the system.
Level 3: Predictive Analytics
We use predictive analytics to do three things.
  1. Classification – classify input data as belong to a specific class. For example, classify the device as an energy hungry device. see https://en.wikipedia.org/wiki/Statistical_classification more information.
  2. Predict the next value – predict the next value in a sequence. For example, predict the electricity demand in the next hour.
  3. Anomaly detection –   find data  points in the data that are different from most data points. e.g Detecting Fraud. This should need time series related anomaly detection as well. See https://en.wikipedia.org/wiki/Anomaly_detection.
Users should be able to select a set of data (e.g. select a subset of devices and a time range) and open them in a Machine Learning Wizard to apply  algorithms and build models. Users should be able to use these models either within other queries or use it to create alerts. New Machine Learning Wizards like Azure ML and upcoming WSO2 Machine learner product are examples of such a wizard (see http://wso2.com/library/blog-post/2015/07/blog-post-sneak-peek-into-wso2-machine-learner-1.0/)