Work-order cycle down 42% — costing closed in 4 days instead of 9.
A trillion-VND ceramics manufacturer brought 3 plants onto one platform: shift output recorded at the station, costing computed per batch, raw-material inventory visible group-wide.
Figures supplied by the customers themselves and published with their consent, measured within the scope and period stated on each metric. Results vary with company size, industry and how standardized the starting data was.
CTH · APODIO is a trillion-VND-revenue ceramics manufacturer with 3 plants and more than 2,000 workers, most of its output exported — which means quotes must be accurate and every batch must ship on spec, on time.
Shift output and scrap were still written on paper at that point and keyed in at end of day; accounting ran on separate software with no link to the shop floor, and planning lived in spreadsheets. Production reports were always a day late, costing took 9 days to close — export orders were quoted against costs from almost two months earlier.
The three plants moved onto an on-premise system pilot first, then rolled out: output and scrap are captured right at the station on tablet/scan, inventory and MRP are unified across all three plants, costing runs per batch with automatic allocation, and shop-floor, warehouse and accounting figures meet on one executive board.
Before and after — with timeframe and method.
| Metric | Before | After | Timeframe & method |
|---|---|---|---|
| Work order → finished-goods receipt cycle | 7.2 days | 4.2 days (−42%) | 3-month rolling average across all orders · after 9 months |
| Monthly costing close | 9 days | 4 days | Per-batch costing · automatic allocation |
| Average raw-material inventory | — | −18% | After 12 months · MRP purchasing tied to the sales plan |
| Output & scrap reporting | T+1 (end of day) | Real-time by shift | Recorded at MES stations across 3 plants |
Figures supplied by the customers themselves and published with their consent, measured within the scope and period stated on each metric. Results vary with company size, industry and how standardized the starting data was.
From discovery to go-live.
Kiln operators were used to paper logs and skipped the data entry. The project fixed the design, not the people: the capture screen was cut to 3 taps, shift leads became the primary recorders, and a per-shift output board went up right on the floor so every crew saw its own numbers. Capture hit 98% by week six and stayed there.
“Costing used to close long after the order had been quoted. Now every batch's cost is sitting in the system — sales quotes on real numbers, not estimates.”
