Cost Center and Allocation Structure

Cost allocations live entirely in the controller's head. When someone asks 'what does it cost to make Product X?' the answer depends on who you ask and what they remember about overhead splits. There is no written allocation methodology — just institutional knowledge about which departments absorb what.

AI cannot perform any cost analysis because no allocation structure exists in any system. Every cost question requires a human to mentally reconstruct the logic.

Someone built a spreadsheet years ago showing cost centers and allocation percentages, but it hasn't been updated since the plant reorganization. The finance team uses it as a starting point and then manually adjusts numbers in their heads. New cost centers get added to the ERP but nobody updates the allocation drivers.

AI could read the spreadsheet but cannot trust allocation percentages because they don't reflect current organizational structure. Any cost analysis based on these rates produces misleading product costs.

Cost centers are defined in the ERP with a consistent hierarchy — plant, department, work center. Allocation drivers (headcount, machine hours) are documented in a policy manual. The controller runs allocations monthly using these rates. But the logic lives in a combination of ERP configuration and offline Excel models that only the controller fully understands.

AI can generate standard cost reports from ERP cost center data, but cannot replicate the full allocation logic because key steps happen in offline models. Product cost accuracy depends on which model version the controller last updated.

The complete cost allocation structure lives in the ERP — cost center hierarchy, allocation drivers, overhead rates, and distribution rules. A finance analyst can query 'show me all overhead allocations to Product Line X for Q3' and get an accurate, traceable answer. Allocation methodology is documented and auditable.

AI can perform full cost variance analysis, identify allocation anomalies, and trace any product cost back to its allocation components. Cannot yet respond to real-time operational changes because allocation rates update on a periodic cycle.

Cost allocation structures are fully schema-driven with formal entity relationships — each cost center links to its parent hierarchy, allocation driver source, consuming entities, and rate calculation methodology. An AI agent can query 'what allocation drivers affect the cost of Product X at Plant 3?' and get a complete, structured answer with full traceability.

AI can perform what-if allocation modeling, recommend allocation methodology changes based on activity patterns, and autonomously flag when allocation rates diverge from actual resource consumption.

Cost allocation structures are dynamic and self-maintaining. Allocation drivers update continuously from operational systems — actual machine hours, real headcount, current square footage. When the plant adds a shift or reorganizes a department, the allocation model adjusts automatically. Product costs reflect reality at any point in time, not last month's averages.

Fully autonomous cost management. AI continuously optimizes allocation methodology, adjusts rates in real-time, and provides instantaneous accurate product costing that reflects the current operational state.

Ceiling of the CMC framework for this dimension.