By Dr. Shi-Wan Lin, IoT Technologist and Co-Chair of the Industrial Internet Consortium Technology Working Group
Industrial Internet is the application of Internet of Things in the industrial space, often referred to as Industrial Internet of Things (IIoT). Industrial Internet systems are undoubtedly diverse and complex systems. To better understand the essence of their development, we can view the development in a few main themes in which these systems are likely to evolve. Here I would like to suggest three main themes: Smart Maintenance, Global Optimization and Local Autonomy. Let me elaborate on each of these themes below.
Broadly speaking, industries at large have made hefty investments in the past decades building up a vast scale of infrastructure for industrial operations (the brown field). This infrastructure consists of various assets, many of which operate at various levels of automation supported by micro-controllers connected to sensors and actuators performing closed-loop controls. Today a large proportion of these assets still operate in isolation, some geographically distributed.
The first phase of many Industrial Internet deployments will likely seek to connect to these assets to remotely monitor and maintain (e.g. configure, update, diagnose, repair, etc.) them. By doing this, we can gain near real-time visibility to the operational states of the assets and an ability to remotely manage them to improve uptime and to reduce maintenance costs. Moreover, data on the operational states from the assets can be analyzed to enable advanced capabilities such as predictive and prognostic maintenance to prevent unplanned downtime and to further reduce maintenance costs through optimal field service planning (over labor, parts and client service scheduling). In doing so, we will gradually gain a higher level of competence and confidence in operating the assets to reliably deliver the desired outcomes. With this heightened level of competence and confidence, we may be able to realize additional cost savings, e.g. by reducing the level of redundancy in the deployment of the assets and by extending the asset service or replacement cycles at least in some cases. Furthermore, we can enhance customer experience by offering stronger service level agreement (SLA), transitioning from a service model of product warranty to that of performance warranty and paving the way to transform the business model from product delivery to outcome delivery.
These are among the much talked about benefits of Smart Maintenance and are recognized as the low-hanging fruits that may provide immediate ROI from the first phase of Industrial Internet deployments. There are many real-world examples of this theme, some already existed and are evolving (e.g. predictive maintenance on aircraft engines, rooftop HVAC systems and commercial coffee making machines) and many more others are being considered and developed.
We need to undertake a few key tasks in order to enable Smart Maintenance. At first, we need to retrofit the existing assets to enable connectivity, data collection and remote management capability. Secondly, we need to establish asset management applications that perform aggregate maintenance on a large number of assets, and streaming analytics and visualization applications to gain visibility to the asset operation in near real time. The analytics will become more advanced progressing from describing the current operation states and conditions of the assets (providing visibility), to predicting how these states and conditions would likely evolve (providing prognostics) and ultimately to prescribing the actions necessary for the continuing delivery of the desired outcomes from the operation of the assets (providing solutions). Finally, we need to integrate these applications to the business systems, e.g. resource and logistics management, work scheduling and customer relation management systems, to streamline the field maintenance services and other business processes that are involved.
We also face a few important challenges. The first challenge is how to implement rigorous security to thwart off potential attacks over the enlarged attack surfaces of the assets now exposed by the connectivity and additional system elements in the network. The second challenge is how to add the new capabilities in a way without compromising the existing operation, its performance and safety. The third challenge is how to create an ecosystem in which data sharing is incentivized so that different vendors of the assets are motivated to share data so that a holistic view and maintenance of the complete system can be realized.
The Industrial Internet Reference Architecture provides a reasonable framework for considering many of these concerns. It also describes some functional and implementation models as a good starting point to conceptualize a system architecture for this theme. Furthermore, a team of security experts from the Security Working Group in the Industrial Internet Consortium is delivering a security framework to address the security challenges. Another team within the Technology Working Group is starting to examine the challenges we are facing in the convergence of Information Technology (IT) and Operational Technology (OT) largely driven by the development of Industrial Internet. The challenges in building Industrial Internet ecosystems to fully realize the potential and benefits of Industrial Internet are also being raised and have generated much interest in the Consortium.
In my next blogs, I will discuss the themes of Global Optimization and Local Autonomy.