By Dr. Shi-Wan Lin, IoT Technologist and Co-Chair of the Industrial Internet Consortium Technology Working Group
This is Part 2 of a 3-part blog series. Read Part 1 here.
With connected assets, we can go beyond the theme of Smart Maintenance -- to seek a higher operational efficiency at the system level, in other words, across fleets of assets. Clearly, after we have worked hard to keep the the assets up and running, we want to know how well they run and how efficiently they operate. By applying analytics on asset level data, we gain operational intelligence from various perspectives, including the operational states, utilization, output and the quality of the output. Based on such operational intelligence, we identify and address any gaps to ensure each of the assets produce the desired outcome efficiently. In many cases, however, we will need to go beyond the notion of efficiency at the individual asset level. We may seek efficiency globally, which may require a coordinated increase of the level of operation at some assets and reduction at others based on business objectives. An optimal global conditional plan might be developed to orchestrate the operations of the assets. The development of such a plan is based on the analysis of the combined operational intelligence and business insights and objectives, taking into account of such factors as the availability of resources, the costs of the operation and the demands of the output. The global plan would then be dispatched to the individual assets to scale up or down their operations dynamically, with updates at increasingly shorter time intervals as technology improves. I see this Global Optimization of asset operations as the second theme of Industrial Internet.
Below I describe a couple of examples to illustrate of this theme.
The first example is concerned with electricity generation. Today, an increasing proportion of generation is based on renewable resources that are typically unsteady and uncertain in supply. To meet this and other new challenges, a power plant operator (a Utility) might build an Industrial Internet system to automatically optimize the electricity generation levels among the set of generation assets under the operator to meet the demand more efficiently. With the new system, a global generation plan within the Utility is optimized dynamically at fine time intervals (it is understood that a certain types of generation assets may take a longer time and higher cost to startup and shutdown thus may be able to change state less frequently). The optimization takes into account of such factors as demand (the total committed generation levels), asset operation condition, energy source (e.g. wind, solar, tide, etc.) availability and its forecast, and generation cost and price for each generation asset type and location. The generation level for each asset based on the global plan is then dispatched to each asset to effect the changes.
The second example involves low cost smart tags with attached sensors (e.g. for measuring temperature and humility, and detecting vibration and shocks). By attaching these tags to objects, we can track the movement of the objects, their condition and environment from parts to products and from factories to shipping and to customers. There are considerable benefits just considering the logistics management of the movement of these objects alone. It could help to optimize the delivery of parts or products by assuring delivery quality (not damaged or spoiled), on-time delivery (optimal routes in response to changing conditions), security (preventing theft), safety (continuing monitoring to prevent hazardous condition and enabling fast response in case of accidents of hazardous material), and lower cost and impacts (optimizing means of shipping). Many of these benefits can only be realized fully with a real time view of the assets (the smart-tagged objects as cargo, and trucks, trains, ships, plans and drones as transport assets) and the ability to globally optimize their movements.
The tasks required to realize the benefits from Global Optimization are complex and challenging. First, we need to identify existing inefficiency in operation, key performance indicators for measuring the efficiency and the goals of improvement. Secondly and more technically, we need to reach a good understanding of the feasibility of measuring the key performance indicators in real time, applying algorithms and rules to determine the best course of actions for optimization and carrying out the actions in orchestrating the operations of the assets, all dynamically. Another task is to integrate these operational processes with business ones allowing the optimization being driven business objectives and constrained by rules of business governance.
The Industrial Internet Reference Architecture (IIRA) can be readily applied to support the Global Optimization theme. For example, the IIRA decomposes an Industrial Internet system into multiple functional domains. The Control Domain concerns about the functions performed by the industrial assets. The Operations Management Domain deals with the functions for maintaining the assets to keep them up and running reliably – the focus of Smart Maintenance discussed in the previous blog. The Information Domain provides functions for gathering data from the assets and from other domains, and transforming them into insights through analytics. The Application Domain focuses on the functions for orchestrating the operations of the assets to achieve optimal performance and high efficiency globally based on the analytic results from the Information Domain – the focus of Global Optimization theme. Finally, the Business Domain consists of business functions to provide business objectives and constraints for the optimization and to offer business support such as resource, logistics, work scheduling and customer relationship management for the smart operation of the assets. The IIRA highlights the importance of integrating operational and business functions in order to realize the full benefits of Industrial Internet Systems.
In the next blog, I will discuss the theme of Local Autonomy.