This audio was created using Microsoft Azure Speech Services
Andrew Hughes joined the LNS Research team in May of 2015 and is a Principal Analyst with his primary focus being research and analysis in the Manufacturing Operations Management (MOM) practice. Andrew has 30 years of experience in manufacturing IT, software research, sales and management across a broad spectrum of manufacturing industries.
In our last blog on Smart Connected Operations, we looked at the trend of connecting new data sources, Smart Connected Devices (SCD), directly to the Industrial Internet of Things (IIoT). It is clear that SCDs are proliferating and will bring new information to manufacturing enterprises. That data is already finding uses in some of the low hanging fruit of the IIoT such as predictive maintenance. However, the promise of the IIoT goes far beyond a few data points gathered by new smart devices.
Thinking Big Data
Big Data Analytics is one of the key promises of IIOT and Smart Manufacturing; while we do not want to
go into depth about the different levels of sophistication in analytics, it is worth considering briefly how we could define big data analytics. I like to use a simple test. Does your system answer questions you did not even know to ask? If yes, you are well on the way to truly smart analytics.
To achieve prescriptive analytics with big data (The top right quadrant), you are going to need to gather data from many sources. Being manufacturing operations people, we tend to think mostly about the manufacturing data. But to get maximum benefit from our IIoT platform, we will want to bring together data from many sources.
As industrial companies continue to drive digital transformation, they want new applications for the ever-increasing volume of data being collected and stored in the IIoT platform. Many applications will take disparate data and use it in new analytical apps, and in ways that were not considered when the data was originally being configured. We need to be able to handle data in new ways, but also to prepare in advance.
https://www.youtube.com/watch?v=GxlYvSZ_a70
Thinking Ahead
As we have discussed, data for any application could come from multiple sources. As soon as you start to combine data from sources provided by multiple vendors, incompatibilities will occur and a bit of planning will go a long way to solving data sharing difficulties. 
As companies move towards digital transformation, data management will be one of the biggest challenges. Everyone from major automation vendors, through a multitude of specialists, to the major business software suppliers, wants to control your data. Indeed, almost everyone has a platform to solve all your integration needs, if we are to believe the marketing hype.
The less easily swayed might think that it is all hype and nothing has changed for decades – we often hear stories that start “We have had IoT for 20 years”. What they mean is “We have had connectivity.”
The reality is that we are at a tipping point. Technology and industrial companies can gain massively by grasping the opportunities. To do so will require making difficult decisions about where data will be managed, and how it will be used. While it is impossible to make specific recommendations today on how to address your data architecture, there are a few key points you should consider:
- Plan big but start small: Consider your mid to long term strategic objectives as the basis for your digital transformation architecture, but start with a small project that has some specific data needs beyond the plant (like integrating customer feedback data).
- Choose a data platform for now: Accept that today’s initial data architecture will not be sufficient for the long term. You may use an existing system such as a MOM database to collect data for an initial project knowing that you will later move most to the cloud or an IIOT platform.
- Ensure you have a sufficiency of data management expertise: As part of building your digital transformation team, everyone wants to bring data scientists on board to do fancy analytics. However, if you cannot manage the incoming and outgoing data, data scientist will not be of much use to you.
Whatever else you do in your digital transformation journey, use data wisely and broadly; easy access to data across your organization will help others and your business. And, at the same time it will help you break down human barriers to success.
To learn more about the impact of digital transformation, watch the video Enterprise Operations Management, read the article Drive Operational Excellence with Digital Transformation, view the webinar Optimize Your Asset Performance Strategy for Maximum Return and explore Schneider Electric’s Manufacturing Operations Management page.
