Acumatica’s 2026 R2 release lets manufacturers record what production actually uses, schedule work to the minute, and ask questions about it in plain language. In a post published October 19, Acumatica presents the three as one flow: the kiosk captures what happens on the floor, Advanced Planning and Scheduling (APS) turns that data into a schedule, and the AI Assistant puts production insights in front of the people who need them.
ERP Today reported on the platform changes at launch: AI agents, Model Context Protocol (MCP) connections to outside AI tools, and a new analytics layer.
Shop Floor Kiosk Records What Production Uses
Acumatica introduced Shop Floor Kiosk in 2026 R1 so production teams could report labor, record completed and scrap quantities, and view materials and recent activity from the floor. R2 adds four updates:
- Operators can issue the exact amount of material an operation requires.
- Operators can issue a material that was not on the original bill of materials.
- Operators can correct a reported quantity on the spot.
- Operators can record lot and serial attributes when finished goods are received.
The first update addresses backflushing, where the system deducts a standard quantity of material for each unit produced. When actual use differs from the standard, the difference goes unrecorded. Acumatica says those differences compound until inventory counts no longer match the shelf, job costs drift from reality, and production margins become harder to protect. The lot and serial attributes support traceability. If a quality issue or warranty claim comes up, a business can search by any recorded attribute and find the affected units.
Acumatica’s post includes a customer view from Andrew Gale, CFO and COO of LiveView Technologies. He says the tighter connection between sales and manufacturing in Acumatica has substantially improved the company’s inventory accuracy.
APS Adds Per-Work-Center Constraints and Minute-Level Scheduling
Advanced Planning and Scheduling (APS) decides when each production operation runs, based on how much capacity the shop has. R2 changes how APS models that capacity.
Each work center can now be set to finite or infinite scheduling. A finite setting limits work to the capacity the work center has, so a true bottleneck is treated as a constraint. An infinite setting lets work centers with spare capacity run without those limits.
Scheduling also runs to the minute instead of in fixed blocks of time, which gives production managers a more precise estimate of when each operation will finish. Overlapping shifts are merged into a single calendar. Acumatica says scheduling has often miscounted that overlap, and the merged calendar means available hours match the people actually working.
Acumatica says APS plans around how long jobs actually take when it is fed accurate labor and material data, and around assumptions when it is not.
AI Assistant Answers Production Questions in Plain Language
The AI Assistant in R2 lets users ask manufacturing questions in plain language. It answers from a set of production inquiries covering material staging, material readiness, performance history, order risk, and schedule status.
Acumatica’s examples are operational. Which inventory items are blocking the most production orders? Which items consistently run over budget? Which orders have the biggest gap between planned and actual cost? Can a given item be promised by next Friday?
Acumatica describes the three features as working together. The kiosk captures what happens on the floor, APS turns that data into a schedule, and the assistant answers questions about material needs, order risk, and delivery dates from the production inquiries.
What This Means for ERP Insiders
Start with the kiosk records. Acumatica positions the kiosk as the point where floor data is captured, so teams that begin with exact-quantity issues and on-the-spot corrections build the record the scheduling and assistant features work from. They gain a clean starting point for everything that follows.
Compare standard deductions with real use before ending backflushing. Checking a few high-volume operations shows where recording actuals would change inventory and costing the most. Teams gain a clear order for rolling out exact-quantity issues.
Set work center capacity and shift coverage deliberately. Finite scheduling treats a work center’s capacity as a limit, so teams that confirm real capacity and shift overlap give APS accurate constraints to schedule against. They gain minute-level schedules that reflect how the shop actually runs.
