Global Cosmetic Manufacturer

In the highly competitive global cosmetics industry — where a handful of multinationals control nearly half of worldwide production — manufacturing efficiency is a survival imperative. When a leading cosmetics manufacturer added a 29th production line at its Dallas plant to increase output of mascaras and lip glosses, the line fell far short of its target throughput of 100–120 containers per minute. Engineers recorded more than 2,500 start/stop cycles per 12-hour shift, yet the plant lacked any manufacturing intelligence capable of pinpointing root causes. Without granular visibility into machine-level events, every stoppage was a lost opportunity and a direct hit to service costs and productivity.

Plant management engaged Prime Controls, a Rockwell Automation Solution Partner for Control, Process and Information, to deploy FactoryTalk Metrics — a manufacturing intelligence software platform from Rockwell Automation. Prime Controls analyzed the new line, which ran as a chain of transfer conveyors across five work-cell stations controlled by Allen-Bradley ControlLogix and CompactLogix PACs communicating over an EtherNet/IP network. Engineers replaced the previous manual fault-selection process with logic that automatically assigns every downtime event to a specific root fault, eliminating unattributed stoppages. The software connects directly to the plant's existing controllers, collects detailed event data across 14 key performance indicators, calculates real-time OEE for each machine, and makes all dashboards accessible via standard web browser — no dedicated terminals required.

Since deployment, the Dallas plant achieved dramatic, measurable improvements in line efficiency:

• 90% reduction in per-shift machine starts/stops — from 2,500 down to approximately 250
• 15–20% increase in overall line uptime
• $100,000 saved annually in service-contract costs and productivity losses

Beyond the headline numbers, plant management gained the ability to evaluate true machine efficiency and product-loss calculations in real time. One early finding — that the bottle feeder was starved for 60 minutes per shift — was resolved through simple restocking. A labeler switching states too rapidly was corrected through a programming adjustment. The solution transformed reactive maintenance into systematic, data-driven process control.

• Visibility precedes improvement: Before optimizing throughput, manufacturers must first establish event-level data collection — you cannot fix what you cannot see.
• Automatic fault attribution is critical: Manual downtime logging creates gaps; logic that auto-assigns stoppages to root causes produces far more reliable OEE data.
• Integration with existing controllers matters: Deploying analytics software that connects natively to in-place PACs and networks avoids costly infrastructure replacement.
• Small programming changes can unlock large gains: Adjusting a labeler's buffer alarm threshold eliminated a recurring starved-state loop with no hardware investment.
• A trusted implementation partner accelerates time-to-value: Prime Controls' engineering expertise was essential for designing fault logic and interpreting initial OEE data.