Quality Loss Reduction at an Aseptic Packaging Plant

How the worst plant in a 40-plant network became the second best, and stayed there

The challenge

Inside a global aseptic packaging company, one site in Brazil had a distinction nobody wanted: worst cost of quality across the entire 40 plant network.

Quality losses were at 9.1%, the result of a bunch of issues. Material was wasted through the process, finished goods were rejected to often, product was returned by customers. It resulted in BRL 20.9M a year in cost due to poor quality.

The plant had a leadership turnover because of the low performance. But that meant each new leader was having to start from a bad spot, without limited institutional knowledge. And the company gave them a hard target: bring quality losses from 9.1% down to 6.4% in 5 months.

The work

We started by mapping roughly 120 process variables across the plant's 8 process steps, then narrowed it down. We found two steps caused the most issues: scraping, with 25 variables, and final folding, with 37. Most of our energy went there.

Eight cross-functional improvement teams took on the losses using PDCA, with 76 employees trained on the job as the work happened. One 5-Why investigation traced a recurring defect to equipment vibration shifting an air nozzle in the card infeed (the kind of thing that would be impossible to find in a conference room.) The team also fixed 30 parameters to prevent settings from drifting between shifts, and created standardized 8 equipment setups with SOPs for each line.

Last, we implemented a four-tier meeting cadence (stand-up, area, plant manager, director) and created a process to review 20+ leading indicators so the plant could spot deviations in real time.

The impact

Five months in, quality losses had dropped to 5.2%, well past the 6.4% goal. Five of the eight improvement teams hit their targets. And the plant climbed from worst in the network to second best among 40+ plants.

Point in time Cost of poor quality (annual)
Start (Aug 2015)BRL 20.9M
Jul 2016BRL 18.6M
Nov 2017 (sustained)BRL 11.8M

The part we're most proud of is that last row. Two years after the engagement the losses were still falling.

Why it worked

We effectively did a Pareto optimization, focusing on the 2 steps that could make the biggest dent.

We trained the team to adopt a new problem-solving culture based on disciplines like the "five why's" and root cause analysis.

And the standardization work helped embed both the solutions we found and a process for finding the next ones into the way the company operated.