When the city of Milwaukee began using a computerized maintenance management system (CMMS) to manage its 50 buildings, operations and maintenance director Joe Jacobsen didn't know how hard it would be to convince other managers of the true measure of his department's performance.
“One thing we do in public works is say, ‘We're going to improve by 10%,'” says Jacobsen, now associate dean of business and information technology at the Milwaukee Area Technical College. “But if we do 25,000 work orders one year and 26,000 the next, that doesn't mean anything. There's no connection to performance. The beauty of these systems is that they can tell us how well the staff can produce—and we can ask them to go there.”
The first thing Jacobsen's team did was analyze a year's worth of preventive-maintenance work orders. “A lot would pop up in one month but very few in another,” he says. “So we took some out of the heavy months and shifted them so that, overall, there were around 200 per month. That way people aren't overworked in one month and sitting around in another month.”
When Jacobsen began considering performance measures, the system moved from the reactive realm to one that could be used to set standards and monitor progress. To create standards for each maintenance task, the department used both the average time and the variances in the length of time it was taking to close work orders to create standards for specific tasks. He then mathematically modeled the system, looking for regions of optimal performance in terms of the time to close a work order as a function of demand maintenance and preventive maintenance.
“This told us which aspects of reaction and proaction were ideal and where the opportunities for improvement existed,” he says.
Though the system proved employees were becoming more efficient, other city managers didn't grasp the significance of a tightening in work-order-closure variances (see chart below).
“When it came to performance measures they actually stood in the way,” he says. “If you identify the optimum performance over the year—the best performance your staff has achieved—and ask them to move toward that goal, you're giving them realistic expectations because they've already done it. Many managers missed this—even made fun of it.
“My whole premise is that we should make decisions based on the data,” he says. “You need to take the data and analyze it.”
Back to main article,"From Preventitive to Proactive"