Connected devices like Sonos and Amazon Echo have become mainstream, and IoT use cases have become one of the most-discussed trends in the last few years. Apart from the consumer facing gadgets and home automation gimmicks, IoT has made some serious impact in Manufacturing, Logistics, Agriculture, Mining, Oil and Utilities.
With the goal of improving efficiency and results, machines and specialized sensors are designed to collect data at every step of the process. This facet of IoT is widely known as the Industrial Internet of Things, or IIoT
Countries like Germany and companies like GE, Cisco, Siemens and IBM have led the IIoT wave. Recently, an increasing number of startups have started to develop sensors, PaaS, and network infrastructure with ML/AI capabilities to extract insights from the deluge of data that invariably arise from all these connected machines communicating with one another and with computers.
History of IoT
In the early 1980s, local programmers would connect to a Coca Cola Machine at Carnegie Mellon University and check to see if there was a drink available and if it was cold, before making the trip. In 2010, Ericsson predicted that there would be about 50 billion connected devices by the end of 2020.
Meanwhile, the manufacturing sector got segmented into discrete and process industries to meet market demands. These industries started to develop operational technologies (OT) like specialized sensors, PLCs, and SCADAs to gather data, control processes, and monitor KPIs
As the competition in the open market grew, the industries needed a way to optimize their processes, and their existing systems were inadequate. Parallelly, the Information Technology (IT) sector was advancing at breakneck pace. With cheap (and powerful) hardware and software, Disruptive ideas came to fruition.
OT systems have not been traditionally networked technologies. Most devices weren’t computerized and those with computing capability were generally closed to proprietary protocols. On the other hand, IT inherently covers communications as a part of its information scope. Thus, the convergence of IT and OT This convergence of automation, communication, and networking in industrial environments is an integral part of the growing Industrial IoT.
Technologies used in IOT
To understand what IIoT is, you need to look at the technologies that are used to build it. Typically, an IIoT system is built with a combination of
1. Sensors
A device that senses the change in electrical signal from a physical device. Recently, these devices have become smaller, inexpensive and robust enough to detect minute changes.
2. Networks
A mechanism for communicating electronic signals. Recently, wireless technologies have evolved enough to deliver 300 Mbps to 1 Gpbs of data.
3. Standards
Technical standards can help systems process the data and allow for interoperability of aggregated data sets.
4. Analytics and BI
Analytical tools that improve the ability to describe, predict and exploit relationships among phenomena.
5. M2M communications
Technologies and techniques that improve compliance with prescribed action.
To wrap our heads around this, we need to look at how all these can fit together to solve a novel problem.
Some Interesting Industrial IoT Use Cases
Predictive Maintenance:
Machines and assets that run continuously have specific maintenance timelines. Outside of these, unplanned maintenance and breakdowns usually cost upwards of $88 million annually. Predictive maintenance can help rein in those numbers.
With the stream of data from sensors and devices, predictive analytics systems can quickly assess current conditions, recognize warning signs, deliver alerts, and trigger maintenance processes.
In an Ideal IoT enabled plant, a motor in the pumping station can schedule a maintenance if it senses anomalies from its sensor data.
This approach promises cost savings over routine and periodic maintenance.
Asset Tracking
The goal of asset tracking is to easily locate and monitor key assets. By tracking assets, industries can optimize logistics, maintain inventory levels, detect inefficiencies or even theft.
Industries like Maritime shipping rely heavily on tracking assets. Specific real-time data, like the temperature of individual cargo, can help the shipping industry save a tonne of money.
Discrete manufacturing industries with distributed assets can leverage IoT to track, monitor, and control assets. For example, an IoT-enabled windmill farm can sense changes in wind speed, direction, and temperature and align individual mills in certain directions and configurations for efficient power generation.
Fleet Management
Trucking and transport is a $700 billion industry. Startups and corporations have started disrupting the industry with autonomous vehicles and IoT technologies. With the right IoT-enabled fleet management tools, the trucking industry can save fuel (by optimising routes), avoid accidents, and improve operational insights through ML and AI.
For example, intelligent pairing of platooning trucks (based on location and route) with Vehicle to Vehicle communication can trucks make coordinated braking and acceleration thus saving fuel (by reducing air drag) and improving safety. Continuous monitoring and tracking can improve operational efficiency by casting a light on inefficient route/behaviour or even time and weather.
Berg insights, a Swedish M2M/IoT research firm predicts 12.7 million active fleet management systems in North America by 2020.
Smart Meters and Smart Cities
Traditional meters measure total consumption, whereas smart meters record consumption against time. This helps utilities companies understand the consumer behavior and reduce operating expenses. Smart Meters can help improve forecasting, streamline power consumption and reduce energy theft.
Making a city “smart” requires
-
Data Driven Urban Planning
Efficient parking facilities, from booking to parking analytics and smart metering.
-
Intelligent Waste Management
Using connected devices to optimize waste collection and recycling process.
-
Location Based Sensors (for monitoring)
Hardware to collect data like current weather, pollution data (Noise, Light, Air and Water).
-
Real Time Tracking and Analytics of Water Resources
Like analyse and test water quality, water level and forecast requirements.
-
Smart Transport and Transit Data
Data-driven commuter shuttle, ride sharing programs, using mobile and sensors to provide analytics for commutes and congestions.
-
Connected Grid and Energy
Analytics and algorithms to manage supply and demand of electricity.
-
Disaster Management
Data-driven decisions for local governments prior to natural disasters.
Current Trends
From Shell to Total to Saudi Aramco, major Oil and gas companies are investing heavily to improve operations and maintenance. Their investments are in IoT technologies like Machine to Machine (M2M) Communication, wireless sensors to analytics software, among others. This widens the scope for IoT use cases.
On the technology side, concepts like Industry 4.0 and Industrial Internet are trending. Though it has been used interchangeably, they are fundamentally different. Industry 4.0 or “Industrie 4.0” came from the German government to promote computerization in the production and manufacturing sectors.
The first industrial revolution was about mechanization; the second gave us mass production, electricity and assembly lines and the third brought us computers and automation. The fourth one, or Industry 4.0, is aimed at bringing connected systems to industries. Industrial internet first came from companies like GE & IBM. These companies came together to set up the nonprofit Industrial Internet Consortium (IIC). It has 170 members from private companies and high pedigree academic institutions. So, essentially one is owned by a government, the other by its many members.
Apart from these concepts, AI in IoT is making some serious strides. There are five places where AI is helping:
-
Data Preparation
Data Lakes can help define pools of data and clean them
-
Data Discovery
Finding a needle in a haystack
-
Visualization of Streaming Data
Lag-less visualization of streaming and discovery data
-
Predictive and Prescriptive Analytics
Predicting problems or prescribing actions from visualized data.
-
Accuracy of Data
Maintaining high accuracy and integrity of data.
Another IoT trend is, Blockchain in IoT. Blockchain is based on 5 main principles like Distributed database, transparency, Peer to Peer transmission, Irreversibility of records and computational logic. By implementing blockchain principles in IoT, Industries can improve workflows, increase safety and visibility, Trustable History/Records and Improve asset lifecycle,
What’s next?
It is not all rosy. IoT is facing some serious challenges. Security concerns like DDoS attacks on IoT devices, connectivity issues, irregular standards and proprietary protocols, talent shortages and a proper business model.