Multi-line MES deployment for a regional food manufacturer — real-time OEE, downtime root-cause capture, quality instrumentation, and ERP integration. OEE moved 18 points in the first year, scrap dropped 22%, and the board got weekly production data on Monday morning instead of a month-end Excel rebuild.
The producer operates six production lines across two sites, making a range of shelf-stable and refrigerated products for domestic and export markets. Production was running 24/7 on lines installed across a decade of organic growth — mix of modern and older equipment, several PLC generations, and operator HMIs that varied by vintage.
Management had three problems. First, nobody could say with confidence what OEE was on any line in any given week — reports were assembled monthly from operator logs and contained substantive errors. Second, downtime root causes were captured on paper tickets that were rarely aggregated. Third, batch genealogy for regulatory or customer-driven recalls required hours of manual tracing across multiple systems.
The operational challenge was not that lines ran badly — skilled operators kept throughput acceptable. The challenge was visibility: without real-time data, improvements were anecdotal and could not be credibly attributed. Management decisions were made on month-old data, and continuous-improvement programmes had no baseline to track progress against.
The technical challenge was that the production environment was heterogeneous — Siemens, Rockwell, Schneider PLCs, different HMI generations, different historian vendors. Any MES deployment had to connect to all of them, in a way the in-house team could maintain long-term.
The engagement started with a two-week plant assessment — our engineers on the shop floor during operating hours, with operators and maintenance, understanding the actual operating reality rather than the formal process documents. The assessment produced a line-by-line instrumentation plan and a phased rollout sequence.
MES selection was deliberate: Ignition was chosen for its strength as a unified OPC UA gateway across heterogeneous control systems — a single platform that could normalise data from every line regardless of PLC vendor. The MES was built to drive a small number of high-value use cases (real-time OEE, downtime capture, batch genealogy) before feature-creep set in.
Rollout was line-by-line over 9 months. Line 1 (the best-performing line) went first to establish the pattern. Line 6 (the oldest) was intentionally last — the most complex integration was left until the pattern was well-established.
"For fifteen years we had operators telling us what was happening on the lines. For the last year, the data has been telling us. Both versions are mostly true, but the data version is more expensive to argue with."
— Plant Director · [VERIFY]Month 1–2: assessment, MES platform selection, architecture design. Month 3–5: Ignition server build, OPC UA gateway to Lines 1–2, first operator HMIs live. Month 6–9: Lines 3–5 instrumented, integration with ERP for production-order and material flow. Month 10–12: Line 6 (legacy), batch-genealogy workflow, quality-data capture. Month 13–14: continuous-improvement handover, RunOps transition.
The most valuable in-programme insight came early: a previously-invisible downtime pattern (brief stops averaging 90 seconds, occurring 20–30 times per shift on Line 2) was identified once MES captured sub-minute data. These short stops had been invisible in operator reporting because no one was asked to report them, but they cumulatively represented 7% of Line 2's available time.
Average OEE across the six lines moved from 61% pre-engagement to 79% at the 12-month mark. The 18-point uplift was driven primarily by availability improvements (reduction in micro-stops and changeover times) rather than performance or quality. Scrap rate fell 22% over the same period, in part because real-time quality data allowed earlier intervention in out-of-spec batches.
The commercial impact was material: additional output capacity of roughly 15% was created without new capital investment. Batch genealogy for audit or recall scenarios was reduced from 4–6 hours to 15–30 minutes. The board report moved from monthly (with retrospective explanations) to weekly (with current data).
Inductive Automation Ignition (MES & SCADA platform), OPC UA gateways to Siemens S7, Rockwell ControlLogix, Schneider M580 PLCs, InfluxDB time-series database, Microsoft Power BI for executive reporting, integration to [VERIFY] ERP via REST APIs, Ignition mobile for operator dashboards.
90 minutes, no slides. A senior partner, your relevant executive, and a whiteboard. We frame the opportunity, the risk envelope, and the engagement shape — then agree whether there's a fit before any proposal is drafted.