Business Intelligence Hits the Shop Floor
In a perfect world, or at least in a perfect plant, there would be not a hint of scrap. Equipment would hum smoothly and schedules would be met like clockwork. But production, in many ways, is the process of uncovering flaws and constantly refining the system.
One tool quickly emerging that could help manufacturers is the increased prevalence of business intelligence software, or BI, as it's commonly called.
On the first day of Invensys' OpsManage'10 customer and partner conference, being held this week in Orlando, Fla., the producer of equipment and controls for automated processes highlighted a handful of companies it is developing products with.
One such partner is a company called Sogeti, which has for the last two years quickly advanced a manufacturing business intelligence system that analyzes every piece of data streaming out of every piece of facility equipment, allowing it to discern trends, such as flaws in a factory line, reasons why equipment is breaking down or even identify questionable material.
The system, called SOMI, which stands for Sogeti Manufacturing Intelligence, has already found a place in the automotive and consumer products markets, and is now founding a home in the aerospace and pharmaceutical industries.
"There's a lot of quality incidents on the market that are happening today and [SOMI] looks at what is causing the issue on the shop floor," said Kristen Schulte, Director of Enterprise Solutions Consulting at Sogeti. "It presents visibility into how the plant is performing and you can get to the root cause of why."
State of the Technology
Business intelligence software has been around since the mid-1990s, but was largely limited to the most lucrative and powerful of companies. Giant retail chains, for instance, used BI to tailor pricing and product offerings down to the store level. Telecommunications companies used it to identify the few thousand customers, among millions, most likely to switch to another cellphone carrier, and to aim marketing at them.
Now the technology is being aimed at a wider audience as a tool to help businesses mine nuggets of insight from mountains of data.
In the case of SOMI, all the information gleaned from a companys manufacturing execution system, such as data from various pieces of equipment, schedule information, process order information is run through a data model and analyzed.
"It's actually performance management," says Schulte. "You are managing how you're performing by correcting a negative trend and replacing it with a positive one. It's also giving visibility on positive trends and what are the manufacturing measures that are coming from the MES system that are producing those positive trends, so you can use those as benchmarks."
It can also be used to measure the progress of a supplier or third party by collecting data off their systems and providing an instant window of performance.
This data explosion allows for more fine-grained analysis, making manufacturing more of a science and, as Schulte describes it, digitizing the enterprise.
"Look, what is keeping CEOs up at night? asks Schulte. "Quality incidents. Well, we're getting to the point where you can take that raw data and have it transformed and into a model where you see your metrics over each shift, over days and over weeks. It's a business measure kept with history so you can see its trend."
