AI Agent Operational Lift for Johnson Screens, A Brand Of Aqseptence Group in New Brighton, Minnesota

Why AI matters at this scale

Johnson Screens, a brand of Aqseptence Group, operates in a specialized niche: designing and manufacturing custom stainless steel screens and filtration systems for water, energy, and industrial markets. Founded in 1904 and headquartered in New Brighton, Minnesota, the company sits in the mid-market sweet spot (201-500 employees) where AI adoption is no longer optional but a competitive necessity. At this scale, the company generates enough structured data—CAD files, engineering specs, order histories, and field performance records—to train meaningful models, yet remains agile enough to implement changes without the bureaucratic inertia of a Fortune 500 firm.

The industrial filtration sector is under increasing pressure to deliver higher reliability, faster lead times, and lower total cost of ownership. Customers in municipal water treatment, oil & gas, and mining cannot afford unplanned downtime. AI offers Johnson Screens a path to differentiate by embedding intelligence into both its manufacturing processes and its installed products.

Three concrete AI opportunities

1. Automated quality inspection with computer vision. Custom screen fabrication involves precise welds and uniform wire spacing. Manual inspection is slow and inconsistent. Deploying high-resolution cameras and deep learning models on the production line can detect micro-defects in real time, reducing scrap rates by an estimated 15-20% and freeing quality engineers for higher-value root-cause analysis. The ROI comes from material savings and reduced rework.

2. Predictive lifespan modeling for installed filters. Johnson Screens' products often operate in harsh, remote environments. By collecting operational data (differential pressure, flow rates, fluid chemistry) and applying survival analysis or gradient-boosted models, the company can predict when a screen will clog or fail. This enables a shift from reactive replacement to condition-based maintenance, creating a recurring service revenue stream and strengthening customer lock-in. A 10% reduction in unplanned field failures translates directly to contract renewals.

3. Generative design for faster quoting. The engineering team spends significant time adapting existing designs to new customer specifications. A generative AI tool trained on the company's CAD library can propose multiple compliant screen geometries in seconds, cutting engineering hours per quote by 30-40%. This accelerates the sales cycle and allows the team to respond to more RFQs without adding headcount.

Deployment risks specific to this size band

Mid-market manufacturers face unique hurdles. First, data often lives in disconnected silos—ERP, CAD, and CRM systems that don't talk to each other. A data integration project must precede any AI initiative. Second, tribal knowledge is concentrated in a few veteran engineers nearing retirement; AI projects must be designed to capture, not alienate, their expertise. Third, the company likely lacks in-house data science talent, so partnering with a specialized consultancy or hiring a single "data product manager" to oversee vendor-built solutions is a pragmatic path. Finally, change management is critical: shop-floor workers and engineers may view AI as a threat. Framing it as an augmentation tool that eliminates drudgery, not jobs, will determine adoption success.