The Changing World: Where IoE Is Making the Biggest Difference
We hope that by now you are able to see how IoE will help you run a more effective business, predict your customer or client behaviors, and automate time and people intensive processes, just to name a few. As you can imagine, this value is realized through a carefully selected and orchestrated ecosystem of elements that produce data, capture all or relevant pieces, and manipulate, correlate, and analyze them to deduce actionable knowledge. The previous sentence is the long way of saying “IoE is about getting knowledge out of information” and, subsequently, of saying “knowledge is power.”
The simplified diagram (Figure 7.1) from International Data Corporation (IDC) gives a very good picture of the stages of data workflow and how data changes from being information created at the process orprocedure level all the way to becoming actionable knowledge.
This rapidly changing world we live in is (and has been for few years) experiencing an explosion of information, and now it is the time to capture and use this information to understand or explain why we did certain things and what we plan to do about certain other things in the future.In the next few sections and chapters, we plan to touch on the preceding statements with real-world examples.
A Shift Toward Insight-Driven Operations
As we bring together all the pieces needed to deliver the highest value from IoE, we must consume and process massive amounts of data from an exponentially growing number of data sources. Often the desired IoE analytics come from secondary and tertiary systems, which require the IoE operator to capture, store, and deliver data to a data management system that manipulates it and makes it available (transmit it) to other systems for further processing, correlation, and analysis. In this case, data manipulation defines data to be processed and presented in a certain way depending on the systems that will perform the final analysis and, subsequently, display results (aka visualization tools). Alongside data manipulation, please note that we also mentioned data capture and data transmission. In the emerging world of IoE, these two concepts are proving to be the most challenging and will be the focus of innovation and research. A simple data capture and transmission scenario is shown in Figure 7.2.
Figure 7.1 Data workflow
Source: IDC Big Data Predictions 2014
Regardless of the market segment or vertical, capturing data related to a particular process or transaction, in most times, may require capturing data related to several other processes and transactions, storing it, and then transmitting it to the data processing engines location for further analysis. Aside from the complexity and expense associated with data transmission, it is inefficient and consumes significant network and computing resources.
Figure 7.2 Simple data capture and transmission scenario
Putting Insight into Action
Imagine that we capture data from a drilling operation of an oil rig with only satellite connectivity. We are talking about hundreds of sensors generating gigabytes of data to be analyzed at the same time. We have to rely on data collected by conventional procedures from pumps, valves, temperature gauges, and the drill-bit itself, stored into historian-type databases, and transmitted to the company’s data center for further analysis. What we see is the emergence of technologies that facilitate the capture of only the data we need from the process or transaction of interest and then transmitting it to the data center for further analysis. This is only a fraction of the data mentioned in the previous example.
In the example, we capture historical data from various sources and send it to a central place for analysis. The data is processed as a batch and the results are possibly presented as a day’s, a week’s, or a month’s operation activities. In the second example or method, the data is collected and processed in real time and as close as possible to the source. A deeper understanding of the business process, the data needed, and the results will help you in architecting your data management and analytics building block (e.g., storage, computing, relational or nonrelational databases, etc.). See Figure 7.3 for an example of data shared outside the boundaries of the enterprise.
Now that we have examined the data capture types and requirements, let us give some thought to data transmission. Capturing data from various levels of operations and transmitting it to a centralized place for processing is not as easy as it sounds. As you probably deduced from our previous satellite transmission example, there are a few important things that need to be considered beyond just the transmission medium. Consider, for example, data security. We have been speaking generically so far, but consider the example of a financial institution or a health care provider capturing customer data or consumer behavior data and transmitting it to a service provider network with a high-performance computing cluster provided by a cloud service in order to be analyzed. Also, consider an example of capturing manufacturing data from your manufacturing floor (or shop floor) where you have your production control systems. This type of environment is usually isolated and controlled even to employees of the same company whose jobs directly relate to manufacturing process. Manufacturers hesitate to allow open communication in their control systems, as any security breach will have data safety and security consequences.
Figure 7.3 Example of data shared outside the boundaries of the enterprise
The next chapter focuses on IoE use cases where early adopters in manufacturing and retail spaces partnered with device makers, IT hardware or software providers, and leading consulting firms to improve business insights and results using IoE enabling technologies.