Proactive maintenance key to plant safety

?Process operators, it seems, are starting to realise the close relationship between safety and the use of predictive maintenance and condition monitoring as part of an effective reliability strategy.

Recent research by ExxonMobil shows that maintenance staff are five times more likely to get injured during a reactive task than during one which has been planned.

Moreover, DuPont - one of the first companies to establish the link between safety and reliability - discovered that of all injuries it had experienced, the most likely was to a maintenance technician with less than two years experience, carrying out a reactive maintenance task.

The view that improved reliability can help achieve safety goals is endorsed by Colin McLean, former general manager of BP Grangemouth, who comments, “A safe site is a reliable site.”

There is strong evidence that plants which are more reliable are also safer, agrees Ron Moore, principal consultant at UK consultancy Reliable Manufacturing.

Moore cites data from a large US paper manufacturing company across some 15 plants worldwide, over a five-year period. This tracked the plants’ asset utilisation against injury rate and found that as asset utilisation goes up, the injury rate came down.

“The correlation between the two is 80%; remarkably high for industrial data,” said Moore. “Most accidents occur in reactive situations. When a plant is running well, with high asset utilisation, there are fewer failures to react to.”

In another example, an Australian chemicals company implemented a more reliable work environment over an eight-year period by increasing the proportion of preventive and predictive maintenance work.

As it did so, there was a clear correlation with the decrease in the number of injuries. The company increased its PM/PdM work from 5,200 orders to 13,250. As a result the total injuries per year more than halved from almost 350 to 160.

Reliable Manufacturing defines reactive maintenance as any work not scheduled at least seven days in advance, said Moore.

“The fewer interruptions for reactive maintenance, the lower the injury rate. Having a weekly scheduled maintenance plan, which gives employees clear visibility of everything that needs to be done in the upcoming week, is fundamental.”

According to Andrew Fraser, MD of Reliable Manufacturing, a key facet of a more reliable and, therefore, safer plant is a reduced reliance of reactive, unplanned maintenance.

However, he said, it is a mistake to think that reliability revolves around maintenance, cautions Fraser. Reliability, he believes, starts with getting the processes right and eliminating existing defects.

50% predictive

“To detect and prevent onset of failure, monitoring the condition of your processes and equipment is essential. In fact, we recommend at least 50% of all maintenance should be predictive.”

For new organisations, or those with new plant or equipment, the focus should then be on incorporating reliability into the initial design, buy and store phases.

“For the majority, of course, plant and equipment are already in place and it isn’t possible to have a significant impact on the first three functions,” said Fraser. In that case, the immediate focus should instead be on installation, start up and operations. In which, condition monitoring has an important role to play.

Fraser believes that condition-based monitoring, and reliability generally, commands a proactive role from operational staff. “After all, who knows better how equipment is changing than its day-to-day operator?

“Using their senses alone to assess process and equipment condition, operators can detect 40%+ of defects, said the Reliable Manufacturing MD. One organisation we work with at RM has only 17% of condition monitoring tasks carried out by maintenance technicians; the other 83% is carried out by operators.”

The advantage of condition-based maintenance is that it doesn’t assume classical failure profile, unlike time-based maintenance, which presupposes all equipment is average and failures will occur according to the statistics available; when this is not likely to be the case.

The implications of predictive maintenance are huge: greater reliability and safety leads to increased asset utilisation and production capability; improved customer satisfaction; and ultimately, revenue growth. “Reliability is not just the safe option, it is the driving force behind a successful organisation,” concluded Fraser.