“The automation products marketplace is commoditized. That happens because the only value proposition on the table is cost,” he says. “We have a commodity—and no one knows what it’s worth.”
How can that be? When vendors ask customers whether vendor-estimated return-on-investment from installing a new automation system was accurate, no straight answers come back, Martin says. “I was shocked to find out that they don’t have a clue.”
What’s responsible for that? Plants’ cost accounting systems, he believes. “They don’t have the relationship established as to how the plant was running before the automation was installed and how [the plant] was running after—to get a true feel for how that [change] was.”
Martin says cost accounting systems provide monthly, plant-wide information. But when an automation system is installed in one section of a plant—“Most people don’t automate the entire plant at one time,” he says—enough things may have happened that management will never figure out the installed automation’s impact. The reason is because the state of the art in accounting hasn’t changed in 250 years, Martin asserts. What he means is that when accounting first evolved around manufacturing, the only practical way to account for what was happening was to pick a period, such as a month, and take a snapshot of the business for that period, which he calls a variance report.
But he wonders why businesses now can’t put part of their accounting processes into automation—and he doesn’t mean voguish key performance indicators, or KPIs. “They are not accounting measures, but operations indicators,” Martin declares. “The fact that you assign money values to KPIs does not make them accounting measures.”
How badly misused are these indicators? KPIs have not proven popular among top level executives, Martin indicates. He’s heard chief financial officers of very large corporations—mining, process and chemicals, for example—say, “If one more engineer provides one more KPI to tell me how much money he made the company, I’ll fire him.”
Fix accounting
So what is the solution? Fix the accounting system, he says. According to Martin, the real issue the accountants have is this: “They want to account for things at the plant-floor level and in real time, but they don’t believe they have a database from which they can extract those data.”
But Martin points out that every manufacturing plant has hundreds of process sensors throughout the entire operation that reveals exactly what occurs every second. “We use that information for process control and alarming. Why don’t we use it for accounting?” he asks.
Connecting software packages, with the hope that such interoperability will produce something good, is not the solution, Martin says. The solution is connecting a distributed control system (DCS) to field systems, and then using that DCS to model and calculate real-time accounting data, such as energy costs, materials costs, yields, contribution margins and the like. “You will have information that has relevance to the accounting system, and thus to the business,” he explains.
However, paramount to justifying automation or making any solution work, is having the right information, Martin affirms. “You must have the right information regardless of the tool—an accounting system, performance-management systems or a dashboard, all which could be used to justify automation.”
C. Kenna Amos, [email protected], is an Automation World contributing editor.
