**This post originally appeared in the SandHill blog**
In looking at Google search-term trends, the business-friendly term “Big Data” has surpassed the more technical term “Hadoop,” indicative of the fact that we’ve finally crossed the chasm from early adopters to the early majority. This, to me, indicates there’s a whole new class of Big Data use cases coming our way in 2014.
The first generation of Big Data use cases were all around understanding user behavior from human-generated data. There are website analytics, understanding where people are clicking and in what order, how long they spend on a site and where they tend to drop off, in an effort to remove any obstacles along the way in the sales conversion funnel. There are credit card transaction data analysis to improve marketing and to find opportunities for cross and upselling. Online gaming companies build entire games around the ability to collect and analyze gamer behavior for iterative game development to keep the gamer engaged longer. And of course there are social media sentiment analysis, network analysis, and more.
This human-generated data analytics, usually with the objective of making more money, is quickly becoming the status quo.
I believe the next generation of Big Data use cases will revolve around making sense of the Internet of Things (IoT) to help optimize manufacturing processes or improve products with the objective of making money by saving time and reducing costs.
A study found that auto manufacturing executives estimate the cost of production downtime ranges from $22,000 per minute to a high of $50,000 per minute, so using data to optimize production is a no-brainer. For example, one major car manufacturer lowered its factory outage time by 15 percent by using Datameer to identify factors that led to robotic failures.
The company combined unstructured and structured data from multiple systems, including PLC (programmable logic controllers) and proprietary factory and maintenance ticketing systems. The PLC devices contained detailed robot data, such as the temperature of components when the robot broke down. To help understand when certain robots broke down in the past, they needed to pull together and analyze temperature and vibration sensor log ﬁles with maintenance history. Using the findings, the company was able to create a robot maintenance schedule to identify and service robots before failure occurred.
There are countless opportunities for the manufacturing business to get in on the benefits of Big Data while it’s still the “early days.” As we’ve already seen in various other vertical industries, I suspect by the end of 2014 we’ll see urgency from manufacturers to investigate and implement Big Data solutions to optimize processes, pull away from their competitors and ultimately save massive amounts of time and money.