Business intelligence (BI) in manufacturing environments is emerging as a worthwhile investment for many savvy companies. These firms are starting to realize that the gap between manufacturing intelligence designed to keep assembly lines running smoothly and their traditional BI systems is more like a chasm.
How does what happens on the production floor affect sales? Profitability? Can small changes driven through many manufacturing sites deliver large returns for the company as a whole?
If companies with significant manufacturing operations want to beam light into the dark, they have to make a case for implementing BI that's capable of creating meaning from manufacturing activity.
Making the case for BI implementation in manufacturing
Building the BI business case begins with a number of questions, including: Where do you start? How do you calculate business intelligence ROI? Who needs to be involved? Who drives the BI projects?
"Depending on the characteristics of the company, who's driving it determines what it [the BI project] looks like," said Simon Jacobson, a research director for Boston-based AMR Research. "You can see anyone from VPs of the supply chain demanding this just for information integration, or you can see this from a chief financial officer who's looking at how to compensate plants, especially throughout manufacturing as a P&L, where the [goal] is to free up as much working capital as possible -- you have to understand how these plants are performing."
Alternatively, a vice president of supply chain might want to understand what performance capacity looks like so he can better map where to drive response based on global demand, Jacobson added.
1. Make sure you have a driver for BI in manufacturing As with any road trip, drivers have the steering wheel, but that doesn't mean they don't get navigational assistance from their passengers. IT may end up taking over the wheel at various points of the project, but organizations need more than just executive-level buy-in. They need executive-level ownership.
"Traditionally, a business guy says, 'I need this type of information,' and an IT guy figures out how to collect and present it," said Boris Evelson, principal analyst of business intelligence for Cambridge, Mass.-based Forrester Research. "IT would be architecting, building, delivering, and buying BI solutions -- that's the traditional approach."
"But that doesn't work," he said. "A business person has to step up and say: 'Not only do I need information, but let me tell you exactly what kind, and let me be the owner of the initiative. It's my budget, my data, it's my neck on the line before my CEO, so I'm going to own the products.' When we hear CEOs, COOs and CMOs talk like that . . . that's when we see successful implementations."
Clearly the person who's driving the initiative has the power to determine the direction and course of action. But organizations should make sure they don't have a driver who's soliciting different destinations from everyone and ends up pulling over in a state of confusion.
2. Find your pain points that BI could solve A company needs to know its pain points and address them quickly during the implementation.
But, at the same time, even if companies fix their existing pain points, they may be ignoring far greater opportunities -- and this knowledge cuts both ways.
"Realize you are going to want to be able to do some analysis on your entire global manufacturing network -- so for a large manufacturer, it's probably a very large project," said Matthew Littlefield, a senior research analyst for Boston-based Aberdeen Group.
And large projects carry expense and risk.
"It is risky in not having a clearly defined goal and not understanding where you're going to derive your value from," Littlefield explained, adding that trying to do too much is a common mistake.
3. Clearly define your goals for BI If an organization doesn't have easily identifiable pain points -- or they don't seem to justify the cost of a manufacturing-focused BI system -- that doesn't necessarily mean there's little to gain. Smart organizations will want to see whether there's information they can deliver to decision-makers that can help them meet company goals or align corporate priorities with business metrics.
The critical IT angle in BI for manufacturing implementations
Meanwhile, even with business-focused stakeholders and clear executive support, IT remains critical in any manufacturing-focused intelligence rollout.
"You need to have a mature IT department within the manufacturers, so you need to have a relatively large size company to have that," Littlefield said.
An underpowered or inexperienced IT staff will have to be addressed right away. In addition, even in an IT department packed with super brains, organizations need to be able to connect three critical groups: IT experts, manufacturing experts and business experts.
"It's tough to do, but we do see that best-in-class manufacturers are considerably more likely do that, and steering committees and cross-functional continuous improvement teams are two ways we see that coming together," Littlefield said.
A steering committee helps direct the vendor selection process and initial project implementation, and once they're up and running, a cross-functional continuous improvement team will ensure that the organization continues delivering new business value.
Understanding the ROI of BI for manufacturers
Determining ROI can be the hardest part of any large-scale business intelligence initiative. Even with pain points and clear goals, business intelligence projects are notorious for sliding out of scope, breaking budgets, and passing deadlines. With large projects that can take months, if not years, a reasonable ROI calculation can be difficult to determine.
Focused BI efforts can often show ROI, but many of those cases also line up with manufacturing intelligence systems that are focused on a portion of a plant's performance -- with very little connection to the broader business. Figuring BI ROI is a difficult proposition, and often a bit of a gamble.
"Can you prove ROI in two years? Nobody can prove that today," Evelson said. "You [must], unfortunately, be a believer that this is the future and you can't survive without it."
He weighs spending $100 million to revamp an organization's BI environment against spending the same amount of money to buy another factor. "Which will give you more ROI? I don't think anybody can come to you and answer this question."
However, savvy manufacturers tend to invest in constant improvement and optimization. That will inevitably lead to a tradeoff down the road when companies that invested in more plants are facing competition from companies that invested in making themselves smarter.
The key is creating short-term goals with long-term investment.
Chris Maxcer is a freelance writer.