Why Manufacturers Should Cash In on the Promise of Business Intelligence

The perceived value of business intelligence (BI) within the manufacturing industry seems low, as industry adoption is slow at best. To bring the value of BI to manufacturers and to identify ways of optimizing a BI environment, I sat down with Robert Abate, the managing principal and global practice lead for RCG Information Technology, a consulting firm that specializes in BI implementations. Robert discussed the ways in which manufacturers should approach BI, including how to optimize current environments and how to avoid common pitfalls when implementing a BI solution.

Lyndsay Wise: What is the best way for manufacturers to optimize their use of business intelligence within the organizations?

Robert Abate: The first thing that comes to mind when talking about optimizing the use of BI is now you're talking about optimizing the optimizer. I would say the best ways for manufacturers to start in business intelligence is to understand what information they're trying to glean from the data, and secondly, to determine the overall goal of business intelligence within the organization. In this regard BI has many roles; it could be the determined supply or the determined demand. It could also be the determined efficiency on the manufacturing floor.

Usually the best thing that I've seen is the use of something called a road map. That is, the analysis of the situation of the specific corporation and the determination of what priorities it should take with regard to the implementation of business intelligence. Many organizations might take a shotgun approach: "Let's buy a tool, let's collect a lot of data and let's see what reports make sense." I often find that an ounce of planning makes for a pound of success. In that regard, taking the time to analyze and architect the solution, which is really engineering a solution, and prioritizing what areas, specifically within business intelligence, would be germane and which areas would bring the highest return on investment.

Consulting organizations are very good at looking at the overall big picture and creating a road map, or a series of actionable projects, that could then implement business intelligence in a phased approach which, [if] trying to implement it as a big bang or boiling an ocean, is a very difficult task. It's often better to take structured or very narrowly focused solutions and build small success after small success. Once a manufacturing executive sees the information and is able to determine actionable steps from it, his or her hunger then increases to say “well, we would like to see this. The next thing I would like to do is this." So the road map or the plan is generally the understanding of what is the strategy of an organization and then turning it into a set of actionable projects or steps.

LW: Why should organizations consider BI, and how does it add value to manufacturing environments?

RA: The value proposition of BI is to improve the organization, improve the efficiencies, and improve capabilities to allow the organization to optimize itself. An organization doesn't just become optimized, it requires management. And the management requires information to make decisions.

Now what is the value proposition? The value proposition is getting the right information to the right person at the right time so they can make the decisions that would then allow the business to adapt. I mean, we're in a very fast-paced economy; add to that global emphasis and global pressures, and what you have is a need for instantaneous decision making, which is maybe not instantaneous, but decisions within one day that could affect the next week or the next month's income for a manufacturer. The need to understand “What is demand?” [and] “What is supply?” is important. If you are deciding to implement BI for the first time, it is important to consider the following issues: Firstly, what is the information needed for your business to be successful? This includes Information in the form of metrics and in the form of data elements.

Secondly, just as critical as business information, which is the reporting of data, is the quality of data. Without a clean supply of oxygen, humans can't survive. Well, without a clean supply of data, in this information age, an organization will fail to thrive. So the need for cleansed information is critical, especially if you're implementing a supply chain and demand planning tool, because if you send garbage in, it comes out multiplied.

And finally, is the need to understand what governance is in place to support the BI initiative you're running. Just as information comes in, information can be good or bad, the quality of the data has importance, so does the ownership of the data. Data coming in from the manufacturing floor, for example, is owned by or generated by that manufacturing floor. So its need to monitor that data, to manage the data and the prophecies associated with governing that data are also critical.

LW: How do organizations avoid these common pitfalls?

RA: The first method is to have a well-defined plan. What I mentioned earlier—a road map. A road map allows you to create an architectural plan, a blueprint, of what you're going to need. An auto manufacturer, for example, couldn't put an engine into a chassis until the chassis was manufactured. So, one of the pitfalls of this is not understanding the key information that an organization needs. I would say one of the ways to avoid that would be to architect a solution, and that means road-mapping it: building the blueprint of what we're going to do, how we're going to do it, and what tools we're going to use. Now, architecture has many areas. There is business architecture. Business architecture is “What is the information I need? How is that information important? What are the metrics? Are there attributes? Do those attributes have dependencies?” Dependencies on attributes are often called business rules. To avoid the first pitfall is to understand that engineering or architecture plays a key role.

After business architecture, we have information architecture, which includes “How do we organize the data? What model do we use? How do we model the data?” Some of the newer trends in architecture use model-driven development, where models are generated either out [of] the data or out of the prophecies, and then that allows us to engineer the solution. Information architecture is highly scalable and supportive of change.

Likewise, there's infrastructure architecture—the hardware and the software that the company decides to invest [to] implement business intelligence. But all stems from, or it all starts with, business architecture: What are our needs? What are the requirements? To avoid the pitfalls, it is important to create the architectural road map, and then make sure you go through the process of requirements, understand what the needs are so that you're architecting the solution to create a solution that supports change.

End of interview

Robert Abate's insights into how manufacturers can build and optimize their use of BI will hopefully help our readers see past the perceived challenges of implementing a BI solution. By adopting a methodology and building a road map that addresses business pains such as supply or demand management, manufacturers can get one step closer to maintaining continual, competitive advantage.

comments powered by Disqus