by Sciforce on Apr 16
Even now, in the post-industrial world, industry remains the biggest requester — and consumer — of progress. In the connected world, it is manufacturing that invests the most in IoT ($189 billion in 2018) as compared to transportation ($85 billion) and utilities ($73 billion) (see IDC Research). Consumer IoT spending are estimated to reach only $62 billion. Similarly, Bain & Company predicts that industrial IoT applications will generate more than $300 billion by 2020, double that of the consumer IoT segment ($150 billion).
It is not surprising that every company tries to implement or claim to implement industrial IoT in their daily practices. At present, IIoT applications connect machines and devices in such industries as oil and gas, utilities and manufacturing.
Industrial IoT vs. other IoT types
Like other types of IoT, Industrial IoT basically uses sensors to collect data that would help speed up processes, gain efficiencies, and ultimately, reduce overall costs for a product or service. At the same time, there are a couple of features that make IIoT distinctly different from other IoT applications:
- the hardware, software and protocols used in industry differ from what we are accustomed to on an enterprise or consumer-grade level. You may never have heard of protocols such as OPC, MQTT DDS or AMQP or SCADA and MES platforms, which are typical for industrial IT unless you work in this industry. The task of IIoT is to connect industrial devices and platforms to applications and data storage mechanisms that are on the enterprise side of the business. At a larger scale, IIoT should be able to merge industrial IT and enterprise IT;
- the physical environments for industrial-connected devices can vary widely, including plants, factories, mines and substations which have been functioning without any IoT for decades.That means installing of sensors in vary harsh conditions with less than ideal, HVAC, ventilation and power capabilities;
- focus on improving efficiency and health and safety as compared to other types of IoT.
However, the word ‘claiming’ is not here for nothing: currently there are over 450 IIoT platforms and without clear definitions it is not always easy to decide what vendors mean. When we consider application enablement platforms, a set of capabilities must be present to distinguish between IoT platforms and software that is labeled as such.
What makes IoT successful?
For example, Frost & Sullivan after comparing IIoT platform vendors and vendor clusters formulated the following key trends in their January 2018 announcement:
- An inclination towards self-service models that would move Application Programming Interface modules to the center of IIoT strategies;
- Open cloud developer platforms which focus on collaboration between IIoT experts and in-house software developers and would potentially accelerate proof-of-concept modeling for customers;
- Satellite-based LPWAN technologies that are expected to overpower cellular-based network technologies such as LTE-M,NB-IoT, and strengthen IoT use-cases for global asset tracking in oil and gas, and transportation;
- AI engines and cognitive capabilities that are about to become a hygiene factor in IIoT platforms, primarily driven by the need for competitive differentiation and better performance.
Yet, the list looks too general and can be applied to almost every sphere. Let’s consider how to ensure that you deal with a real IoT platform that would make deployments successful and efficient.
At first, let’s look at general characteristics of an IoT platform:
In terms of data management, it is important to understand, that the amount of connected devices can reach hundreds of thousands and in some industries, they can be scattered across vast areas. Once such devices are deployed in the field, it’s very costly to maintain them manually. An IoT platform with full device management capabilities will enable an enterprise to centrally manage the entire fleet of IoT devices through a single panel.
Device communication management
A greater number of devices usually means, on the one hand, more data incoming and, on the other hand, more risks of communication ruptures due to instability of connection and different origin and type of connected devices. From this perspective, the IoT platform should meet the following requirements:
From general IoT to specific industrial IoT features
Besides from the above mentioned characteristics that make an IoT platform effective, in the industrial sphere, there exist a number of specific features and limitations for IIoT systems:
Data collection and analysis: each industrial IoT deployment has unique characteristics in terms of the data management and scale of deployment. The types of data it uses, collection methods, and analytics likely come from sensors and software deployed within highly complex and customized machinery.
Challenging environments: in many cases, sensors and other sensitive network equipment for industrial IoT are deployed in areas with harsh environments, including high heat, extreme cold, high humidity, and poor ventilation, requiring special ruggedized IoT sensors and network hardware.
Quality control: one of the key peculiarity of industrial IoT is that much of the data collected for IIoT deals with automating quality control processes.
Energy efficiencies: discovering and automating energy efficiencies is one of the crucial tasks of IIoT which involves the implementation and collection of smart sensor data in HVAC and lighting systems, implementing artificial intelligence into plant equipment to keep them running at optimal levels, and targeted modernization of plant to enhance energy efficiency.
Improved supply chain visibility: many industrial IoT projects are designed to provide end-to-end supply chain visibility avoiding blind spots that are typical for supply chains equipped with built-in digitized silos.
Retrofitting equipment: industrial IoT projects often involve retrofitting industrial equipment, including manufacturing equipment, forklifts, and storage containers, with IoT sensors. Most such projects require updating legacy processes and tools to bring them into the world of IIoT.
Therefore, to provide a viable IIoT platform, vendors and developers should ensure modularity, interoperability and an open approach on the level of supported cloud service providers and offer companies and IIoT platform that would grow with them.
Though the IoT platform market is far from mature, and not all vendors offer IIoT solutions in the strict sense, still there are clear innovators, working for instance in the scope of edge and/or fog computing and paving the way for Industry 4.0.