Purchase Requisition

Purchase requisitions are not defined consistently. When someone needs supplies, they email their manager with what they want. Different departments use different terminology — operations asks for "pallet jacks," maintenance requests "material handling equipment," and warehouse says they need "new lift equipment." There's no standard template or required information. Managers don't know if a request should include supplier preferences, urgency, or budget codes.

None — AI cannot generate procurement requests because there's no standardized requisition format to follow, no defined fields for automation to populate, and no clear rules about what information belongs in a requisition.

Purchase requisitions follow a basic template but many fields are optional and usage varies by department. Everyone uses the same Word form, but operations fills out "urgency" while warehouse skips it. Some people put supplier names in "notes," others leave it blank. The procurement team has informal rules — requisitions under $1,000 don't need budget codes, safety equipment needs manager approval — but these aren't documented in the template itself.

AI could read requisition templates but inconsistent field usage and undocumented rules mean automated requisition generation would miss critical information or violate approval policies. Inventory-triggered requisitions would work for simple items but fail for anything requiring special handling.

Purchase requisitions use standardized fields and approval rules that all departments follow. Every requisition includes item description, quantity, unit cost, total cost, delivery date, cost center, and business justification. The system enforces data quality — you can't submit a requisition without a valid cost center code. Approval workflows are documented: under $500 requires supervisor approval, $500-$5,000 needs department manager, over $5,000 requires director sign-off.

AI can generate basic requisitions using standardized fields and route them through defined approval workflows. Inventory systems can automatically create requisitions when stock hits reorder points. However, AI cannot optimize procurement strategy because requisition-level standards don't connect to supplier performance data or broader spend analytics.

Purchase requisitions are part of an integrated procurement framework. Each requisition links to strategic sourcing policies — when creating a requisition for diesel fuel, the system references the fuel supply agreement and pre-fills the contracted supplier and pricing. Requisitions for maintenance parts automatically check if items are covered under service contracts. The approval workflow considers not just dollar thresholds but also strategic importance — requisitions for sole-source suppliers require additional review regardless of amount.

AI can generate intelligent requisitions that leverage strategic sourcing relationships and contract terms. The system automatically optimizes supplier selection based on contracts, pricing, and performance history. However, AI cannot adjust procurement strategy in real-time because requisition standards are updated quarterly rather than continuously.

Purchase requisitions operate within a dynamic procurement framework that updates continuously. When a carrier's on-time performance drops below 95%, requisitions for their services automatically flag for review. If diesel prices spike 15% above the contracted rate, fuel requisitions route to the procurement director for supplier renegotiation. The system maintains real-time supplier scorecards, market pricing indices, and contract compliance metrics that inform requisition standards as conditions change.

AI has complete autonomy in requisition generation and procurement optimization. The system automatically adjusts supplier selection, initiates contract renegotiations when market conditions warrant, and optimizes requisition timing based on pricing trends. Fully autonomous procurement operates within dynamically updated standards.

Purchase requisitions operate within a self-optimizing procurement framework. The AI continuously experiments with supplier selection strategies, learns from every procurement outcome, and automatically refines requisition standards. When the system notices that requisitions processed before 10 AM convert to purchase orders 30% faster, it adjusts inventory trigger timings. After detecting that requisitions with detailed specifications receive better pricing, it automatically enhances requisition templates. The framework evolves itself based on procurement performance data.

Fully autonomous, continuously learning procurement. The system optimizes not just individual requisitions but the entire procurement process architecture. AI automatically identifies inefficiencies, tests solutions, and implements improvements to requisition standards without human intervention.

Ceiling of the CMC framework for this dimension.