Turbocharging equipment with new intelligence

Magazine Article, Source : The Manufacturer US
Zone : Operations and maintenance

It’s tempting to upgrade older equipment, finds John Harney, but Coca-Cola Amatil saved $1 million and produced a half million more cases of Coke through sheer intelligence

A uniquely configured production line can do what no other production line can, and thus it can distinguish its owner company from competitors. That said, it presents vulnerability. Spare parts are not around the corner, not even necessarily on the shelf. In an industry that measures throughput per minute, that downtime is murderous to profit.

Now put that production line on the other side of the earth. There you have the unique challenge of Coca-Cola Amatil, the bottler of Coke and other beverages in Australia and New Zealand. Its primary plant is in New South Wales, Australia, where 120 employees produce millions of containers of beverage a day on four production lines. Those four lines are tightly configured, with little room to reconfigure, says Wayne Luxford, the plant’s continuous improvement manager. On top of those challenges, the plant was as large as it could get on its site, which would not allow for new production lines for added capacity. So Luxford was charged with higher throughput out of existing equipment that was up to 30 years old, without the luxury of a lengthy implementation or shutdown.

Luxford saw this as a matter of brains versus brawn: the equipment, he felt, was capable of far higher capacity. But until 2003, when Coca-Cola Amatil installed manufacturing intelligence (MI) software from Activplant on all the lines, “We were relying on pen and paper for guys to write down [production line] throughput and downtime,” says Luxford, “so keeping track of stock leaving the plant was not as accurate as it could be.”

As well as creating confusion about how much stock was produced and then warehoused, manual counting was slowing production. “We weren’t able to meet our scheduled demand for what our sales guys wanted in the warehouse,” says Luxford. However, the company didn’t have the time, money, or space to build a new line to up production.

The Activplant MI software is comprised of a server that taps into the Program Logic Control (PLC) of line machines to collect whatever data plant personnel choose, such as minutes of uptime, case counts, and alarms. That data is stored in a database on the server and presented in a web browser. Luxford says the data is sliced and diced so it’s easy to see the overall performance of the line and each machine in it.

Since the software can monitor anything that has a PLC signal, says Luxford, the company has been measuring, in addition to cases and bottles off the production line to the warehouse, how much CO2 and water it uses. It also measures the overall kilowatts of power each line uses, which has led the company to install power-saving devices throughout the plant.

One of the key benefits of the MI software is identifying alarms that can shut down an asset and create bottlenecks. Says Luxford, “We’ve found a lot a value in tracking alarms that cause the assets to stop. Because of the way the database is configured, I’m automatically capturing the PLC alarms that stop the assets. This way we can see what alarms happen a lot or for a long time, and then we investigate to work out how to stop that alarm from appearing.” For instance, by monitoring the software data output he can see if a labeler can’t run for some reason, which affects the productivity of the next asset down the line. Rather than relying on empirical data, personnel previously had to constantly observe the line and try to identify causes for these types of underperformance.

With Activplant, says Luxford, plant personnel obviously “know a lot more about their lines. They can see their output data almost minute to minute. They can see how production is going from hour to hour, which is key. A really good day for us on our can line is a million cans in 12 hours. They have a production view, so they can see in hourly chunks how many cans they’ve filled and see what the total is for the shift thus far, and that gives them a good idea of how they’re running.”

The company’s stock losses are less too. “We have a measure called IRA (Inventory Record Accuracy), and that used to be quite low. Now that we’ve got a precise count of what we’ve made, our IRA is very good because we’re not losing stock. In the past we’d be posting production numbers that weren’t accurate, so the warehouse guys were looking for stock that was never actually made. Now we can look at the last shift and figure it out. We can even look at what came in as raw materials and what was put into the warehouse and figure out accuracy from that.”

Activplant also comes with the ability to add notes to data collected from plant assets. So, says Luxford, “When a line does stop, we can add commentary about why the line stopped.” This informs the next shift about the problem so they can monitor it.

Luxford has also arranged for production, electrical, and engineering personnel from the day and night shifts to meet once a week at shift change so he can summarize how all the lines are performing, indicate which machines are causing problems, and proactively schedule maintenance on assets requiring it so the line is not interrupted unnecessarily. For example, one conveyor jammed up to 150 times a day—quickly fixed by a line operator, but still significantly reducing output. The team recommended a slight reconfiguration that reduced the jams to less than five per day.

In addition to the other benefits, Luxford says, “We’re running faster now, so we’re making more product in less time than before.” He estimates this saves the company about

$1 million a year, for an ROI of over 800 percent since the software was installed. And the plant has increased throughput by 500,000 cases a year without higher labor costs and on that older but clearly rugged equipment.

“Absolutely,” says Luxford, when asked if a company seeking higher throughput should look to systems first before equipment. Not that Coca-Cola Amatil does not maintain and upgrade that equipment, but as Luxford describes, Australian manufacturers are used to finding a way to keep going in the event of a breakdown and with a lack of spares. The company is looking into across-the-board equipment upgrades, but for the time being it is so delighted with the results at the New South Wales plant that it is implementing the same system at six other plants.

Comments on this story

click here to add a comment

Loading