Unit 3

Industrial IoT (IIoT)

Definition
• Industrial IoT (IIoT) refers to the network of connected devices, sensors, and machines used in
industrial settings to collect, exchange, and analyze data for enhanced operational efficiency,
safety, and productivity.
• Essentially, it extends the concept of the Internet of Things (IoT) to manufacturing, energy,
transportation, and other industrial sectors, enabling real-time monitoring, predictive
maintenance, and automation.
• It is the application of Internet of Things (IoT) technologies within industrial sectors to enhance
operational processes, efficiency, and decision-making.
 Applications:
• Predictive Maintenance: Detecting equipment anomalies before failure, reducing downtime.
• Smart Manufacturing: Real-time monitoring of production lines (Industry 4.0).
• Energy Management: Optimizing grid operations in utilities.
• Supply Chain Optimization: Tracking assets and inventory in logistics.
• Agricultural Automation: Precision farming using soil and weather sensors.
Definition (conti.)
 Key Aspects of IIoT:
• Data Collection and Analysis: Sensors and smart devices capture data from machinery and
processes, which is then analyzed to identify trends, detect issues early, and optimize
performance.
• Connectivity: Robust network infrastructure, including wired and wireless communication, is
critical for the seamless transmission of data between devices and central systems.
• Automation and Control: IIoT enables automated decision-making and control in processes,
reducing human error and increasing operational efficiency.
• Predictive Maintenance: By continuously monitoring equipment conditions, IIoT systems can
predict when maintenance is needed, helping to avoid costly downtime.
• Enhanced Safety and Efficiency: Real-time monitoring of industrial environments improves
workplace safety and streamlines processes, leading to energy savings and reduced
operational costs.
IIoT Business models
• IIoT business models are frameworks that enable industries to monetize connected devices, data,
and services by leveraging real-time insights from industrial equipment.
• These models transform traditional industrial operations into data-driven ecosystems,
emphasizing efficiency, scalability, and customer-centric solutions.
 Key IIoT business model:
1. Subscription/Service-Based Models:
• Customers pay recurring fees (monthly/yearly) for access to IIoT-enabled services.
• his model ensures steady revenue streams and fosters long-term customer relationships.
• Example: BMW offers connected vehicle services, including real-time diagnostics and
over-the-air updates to 20 million vehicle.
IIoT Business models (conti.)
 Key IIoT business model (conti.):
2. Pay-Per-Use/Outcome-Based Models:
• Customers pay based on actual usage or outcomes achieved.
• Example: Companies charging airlines based on engine usage hours monitored via IoT
sensors.
3. Asset-Sharing/Leasing Models:
• High-cost industrial assets (e.g., machinery, drones) are shared among multiple users,
maximizing utilization and reducing idle time.
• Companies leases connected agricultural equipment, enabling farmers to pay only for
operational hours.
4. Data Monetization:
• Selling aggregated, anonymized data from IIoT devices to third parties (e.g., insurance
company)
IIoT Business models (conti.)
 Key IIoT business model (conti.):
5. Servitization (Equipment-as-a-Service, EaaS)
• Transitioning from selling equipment to offering it as a service, bundled with
maintenance and upgrades.
• Example: Companies provides software based industrial analytics and optimization tools
via subscription
6. Ecosystem Platforms:
• Creating IIoT platforms that integrate partners (e.g., suppliers, developers) to deliver
end-to-end solutions.
 Key Challenges:
• Cybersecurity: Protecting sensitive industrial data from breaches.
• Interoperability: Integrating legacy (older) systems with modern IIoT platforms.
• Organizational Change: Aligning IT/OT teams and upskilling workforces.
IIoT Architecture
• IoT architecture is the structured framework
that defines how various IoT elements (devices,
networks, sensors, applications) interact to
form a complete IoT system, encompassing
layers, components, and data flow from devices
to applications.
• The solution gathers data from a large number
of devices, which is then sent to data centers
and servers for further analysis which drives
automation and actions.
• However, between your command and the
achievement of your tasks, there remains an
extensive and largely invisible architectural
framework that relies on multiple elements and
interactions.
IIoT Architecture (conti.)
• There are four layers present i.e., the Perception Layer, Network Layer, Processing Layer, and
Application Layer.
• Perception/Sensing Layer:
• The first layer of any IoT system involves “things” or endpoint devices that serve as a conduit
between the physical and the digital worlds.
• Perception refers to the physical layer, which includes sensors and actuators that are capable
of collecting, accepting, and processing data over the network.
• Network Layer:
• Network layers provide an overview of how data is moved throughout the application. This
layer contains Data Acquiring Systems (DAS) and Internet/Network gateways.
• DAS performs data aggregation and conversion functions (collecting and aggregating data
from sensors, then converting analog data to digital data, etc.
IIoT Architecture (conti.)
• Processing Layer:
• The processing layer is the brain of the IoT ecosystem. Typically, data is analyzed, pre-
processed, and stored here before being sent to the data center. This is where Edge IT or
edge analytics enters the picture.
• Application Layer:
• User interaction takes place at the application layer, which delivers application-specific
services to the user. An example might be a smart home application where users can turn on
a coffee maker by tapping a button in an app.
• There are many ways in which the Internet of Things can be deployed such as smart cities,
smart homes, and smart health. If it is deployed for industrial purpose, it is called as
Industrial Internet of Thing.
Stages of IoT Solutions Architecture
• Having discussed the IoT layers, how can businesses benefit from them and how can they
maximize the value of IoT?
• The Internet of Things (IoT) may refer to connected devices and protocols, but in reality, the data
from these devices is siloed, fragmented, and isolated.
• As such, these fragmented insights alone do not provide enough information to justify an IoT
strategy that involves a significant investment of resources.
• To capitalize on IoT, enterprises must allow devices to interact freely, and they must maximize
device and system synergies. You need to ensure your infrastructure supports the IoT
architecture.
• The following are various stages of IoT architecture implementation in enterprises (next page):
Stages of IoT Solutions Architecture
(conti.)
• Connected Objects/Devices:
• As a first step towards IoT architecture, the physical layer must be established within the
environment. There would be no Internet of Things without “smart” or connected objects.
Typically, these are wireless sensors or actuators in the perception layer.
• Sensors collect and analyze data from the environment in order to make it usable for
further analysis. Actuators are involved in measuring the change recorded by the sensors.
• It is possible to connect sensors or actuators in a wired or wireless manner in order to
perform sensing and actuation. Local Area Networks (LANs) and Personal Area Networks
(PANs) can be used for connecting sensors and actuators.
Stages of IoT Solutions Architecture
(conti.)
• Internet Gateway:
• When step one is done properly, the next step that needs to be done is to set up an internet
gateway.
• As the sensors and actuators collect data in analog form, we must have a means of
converting the analog data into digital data in order to process it.
• In the internet gateway stage, raw data will be received from the devices and pre-processed
before being sent to the cloud.
• Data Acquisition Systems (DAS) can be used to convert analog data into digital forms. It
connects to the sensors and actuators and gathers all data, converting it to digital form so
that it may be routed over the network by the internet gateway. It is responsible for data
aggregation and conversion.
Stages of IoT Solutions Architecture
(conti.)
• Edge IT Systems:
• The third stage of an IoT architecture involves pre-processing and enhanced data analytics.
• Edge IT systems play a crucial role in reducing the pressure on the core IT infrastructure.
• Edge IT systems employ machine learning and visualization techniques to generate insights
from collected data.
• Machine learning algorithms provide insights into the data while visualization techniques
present the data in an easy-to-understand manner.
• Data Centers and Cloud Storage:
• After the data has been properly preprocessed and analyzed, and all loopholes have been
removed, the data is sent to the data centers and servers for final analysis and reporting.
• Data Centers and Cloud services fall under the Management Services category and usually
process data through analytics, device management, and security controls. Furthermore, the
cloud also enables the transfer of data to end-user applications like Healthcare, Retail,
Environment, Emergency, Energy, etc.
Industrial IoT Sensing
• Industrial IoT (IIoT) Sensing refers to the use of advanced sensors.
• These sensors act as the "eyes and ears" of IIoT systems, enabling industries to optimize operations.
 The most popular types of industrial IoT sensors are:
• Temperature Sensors: Temperature sensors, as the name suggests, are used to ensure the
temperature of a device.
• Humidity Sensors: These types of IIoT monitoring sensors often go hand-in-hand with the
previously discussed temperature ones. Primarily, because humidity also has an enormous impact
on the quality of certain food products.
• Vibration Sensors: In the industrial context, vibration sensors are crucial. In particular, due to the
fact that irregular vibrations often come as a precursor to machinery and equipment failure.
Vibration sensors collect readings which are then analyzed to detect if they fall outside typical
specifications. If so, it could be a sign that a machine isn’t running properly and an issue could arise
soon.
Industrial IoT Sensing (conti.)
 The most popular types of industrial IoT sensors are (conti.):
• Proximity Sensors: This category of IIoT sensors is used by manufacturing or other industrial
companies to accurately measure the distance between objects. Typically, these tools work by
emitting electromagnetic fields.
• Gas Sensors: It’s no surprise that a gas leak can pose immense danger to the employees working
in your facility. Plus, of course, it can place the entire business in jeopardy if the problem is not
contained in a timely manner. So, many manufacturers, energy companies, and other industrial
players choose to implement gas sensors when considering the development of Internet of Things
solutions.
• Current Monitoring Sensors: There are two functions that current monitoring sensors can deliver.
First, they provide accurate power consumption readings that help keep utility bills under control
and facilitate sustainable operations. However, that’s just the beginning. An arguably even more
useful benefit of these devices lies in their ability to forecast machine failure. You see, just like
odd vibrations can be a foreshadowing of an issue, so can sudden electric current spikes or drops.
Industrial IoT Sensing (conti.)
 The most popular types of industrial IoT sensors are (conti.):
• Security Sensors: Security sensors are used to detect trespassers and unwanted visitors.
• Level Sensors: Level sensors are usually installed to monitor levels of fluid, powder, or other
materials within a piece of equipment or a container.
• Pressure Sensors: These sensors are used to observe pressure changes in gasses and liquids that
your industrial facility may be working with.
• Speed Sensors: For monitoring speed.
• Infrared Sensors: Infrared sensors can assess the surrounding environment through the emission
of infrared radiation and the detection of it. They can also be prevalent in wireless tools that
require a remote-controlling function. The truth is these devices have quite a wide range of uses,
including serving as item counters, burglar alarms, radiation thermometers, and even gas
analyzers. In fact, some of the tools we’ve discussed above are combined with infrared sensors for
an even better performance.
Industrial IoT Sensing (conti.)
 The most popular types of industrial IoT sensors are (conti.):
• Anti-Theft Sensors: Anti-theft sensors are used precisely for what the name suggests — theft
prevention. Retail tech solutions often rely on them to ensure items stay within a permitted area,
typically the store.
• Air Quality Sensors: Another industrial IIoT sensor has everything to do with employee wellbeing.
Air quality sensors are used to maintain healthy work environments for factory or production
plant staff as these facilities can produce harmful fumes and toxins.
• Chemical Sensors: This category of IIoT sensors is designed to detect and measure the presence
and concentration of chemical substances in various environments.
• Fire Detection Sensors: This technology includes smoke detectors, which sense airborne particles,
heat detectors that respond to temperature increases, and flame detectors that identify light
wavelengths specific to flames.
Data management with Hadoop for
IIot
• Hadoop, an open-source framework, is a powerful tool for managing the vast amounts of data
generated by Industrial Internet of Things (IIoT) devices, enabling efficient storage, processing,
and analysis for actionable insights, and optimized operations.
• Hadoop is designed to handle big data, offering a scalable and distributed solution for managing
and processing this data.
• It process large data sets in a parallel and distributed manners.
• Hadoop framework uses python, java, C++ codes.
• Hadoop consists of two primary components:
• HDFS (Hadoop Distributed File System): For distributed storage.
• MapReduce: For distributed processing (though modern Hadoop often uses YARN for
resource management and supports other processing engines like Spark)
Data management with Hadoop for
IIot
• Working:
• Splitting Data: When you upload a file (e.g., sensor data from IIoT devices), HDFS splits it into
blocks (default size: 128 MB or 256 MB). Example: A 1 GB file is split into 8 blocks (if block
size = 128 MB).
• Distributed Storage: Blocks are stored across multiple DataNodes (worker nodes) in the
cluster. Each block is replicated (default: 3 copies) across different nodes to ensure fault
tolerance. The NameNode (master node) tracks where blocks are stored (metadata
management: location, file size, and status of each block). Splitting and distribution of data is
performed by HDFS.
• Processed (MapReduce): Hadoop processes data in parallel using a "divide and conquer"
approach. The input dataset (e.g., sensor logs) is divided into splits. Each mapper processes a
split of data, transforming it into intermediate key-value pairs. Intermediate key-value pairs
are grouped by key and sorted.