项目背景
Hubei Yuenong Machinery Co., Ltd. builds combine headers, auger troughs and grain handling attachments in Zhumadian, Henan. The plant ships into a domestic dealer network and into export programmes in Central Asia, producing roughly 6,000 wear-plate assemblies a year in a mix of about 400 different part numbers.
Wear parts — cutting blades, auger flights, sieve frames and impeller liners — were the hardest part of the mix. HB400 plate between 8 and 25 mm was cut on a CNC plasma table, and the plates carried heavy dross that had to be chipped off by hand and ground before welding. On an average flight plate that was 18 minutes of grinding per part.
Simpler wear parts went to two outside laser shops 150 km away, at 30% of the part value, and the plant absorbed a fortnight of transport on every order.
The business had outgrown its wear-part cell: two plasma tables were running two shifts and still held a 12-day backlog before the harvest season.
客户挑战
HB400 at 25 mm was the limit of the plasma table. Anything above that went to the outside shops, which meant the two highest-value wear parts in the programme were never produced under the plant's own quality system.
The dross-and-grind sequence was the real cost. On a flight plate, 18 minutes of grinding was added to a 9-minute part, and the grinder was the constraint on the whole sub-assembly line.
Seasonality is brutal. Pre-harvest orders ran 60% above the previous year, and the response was always overtime and temporary labour, which the plant wanted to stop doing — it was a common source of scrap errors on HB400 edges.
Two plasma tables also meant two consumable gas lines, two sets of spares and constant troubleshooting of arc wander on thick plate.
解决方案
Yuenong installed an LP-F6025-20kW Heavy-Duty fibre laser in the wear-part area, where the 20 kW source cuts 25 mm HB400 in a single pass without the multi-pass behaviour that was eating the plasma table's consumables.
An LP-F3015-6kW was installed in the general fabrication bay for the 6–12 mm mix, which covers most wear-part and structural part numbers and gives the shop a fast, low-cost machine for the high-volume items.
Both machines draw from one shared programme library and one nesting engine, so the 6 kW takes the small orders and the 20 kW takes the thick work without a second planning routine. Thick-plate jobs are routed automatically by the nesting engine rather than by a planner's judgement.
The manual grinding station was replaced with a light deburring pass — enough to remove an occasional thermal spot, but not a repair operation — and the remaining grinding moved downstream of welding, where the geometry is better supported.
Seasonal peaks are now handled by running the 20 kW machine a second shift for six weeks, and the cell is staffed by one operator rather than a two-person plasma team.
交付成果
The average cut cycle on HB400 wear plate fell by 56%, measured over a quarter of production data on identical part numbers. Because the plate no longer needed multi-pass cutting, gas and consumable cost per kilogram dropped 43% and the maintenance burden on the wear-part cell essentially disappeared.
Grinding on wear parts fell from 18 minutes per flight plate to under 3 minutes, freeing two grinders that were moved into the welding line. The pre-harvest backlog that had stood at 12 days was cleared in three weeks of the first peak.
Roughly 70% of the wear-part volume that had been placed with the two outside shops came back in-house, and both shops were dropped from the supplier base. The plant recovered the higher margin on those parts and the fortnight of transport risk that came with them.
Material yield on the wear-part programme rose by 5 percentage points once thickness groups were nested independently rather than on a shared plasma programme.
First-pass yield on wear-part assemblies, measured at final inspection, moved from 94.1% to 98.4% over two quarters, with the improvement concentrated on edge-quality rejects.