The FourKites Knowledge Center

Service-as-Software is the commercial model in which AI performs actual work (resolves exceptions, processes documents, schedules appointments) rather than providing tools for humans to perform the work, representing the shift from selling software access to selling completed services.

Every enterprise technology leader faces the build-vs-buy question for supply chain AI agents. The prototype is seductive: a working agent in a week using LangChain or CrewAI. This framework examines five problems that emerge between demo and production, and provides criteria for when to build internally versus when to buy a platform.

Agent Communication Channels are the modalities through which supply chain AI agents reach carriers, suppliers, customers, and enterprise systems: Email, SMS, System-to-System API, Portal/UI, EDI, and Voice.

EDI 214 (Transportation Carrier Shipment Status Message) and EDI 990 (Response to a Load Tender) are standard electronic data interchange formats used for carrier-shipper communication. EDI 214 transmits shipment status updates (picked up, in transit, delivered). EDI 990 transmits carrier acceptance or rejection of a load tender.

AI workflow automation is the use of artificial intelligence to execute multi-step business processes autonomously, replacing manual task sequences with agent-driven workflows that gather data, reason on intelligence, take action, and record outcomes without human intervention for routine operations.

FourSight AI is the intelligence and analytics layer of the FourKites platform, providing conversational analytics, generative interfaces (Gen UI), automated scorecards, and the bridge from insight to autonomous action.

The Consultant Tax is the recurring cost (typically $500K-$2M per implementation cycle) that enterprises pay to systems integrators and consulting firms to configure, customize, and maintain supply chain automation that could be built by an AI Developer Agent in days.

The FourKites Graph is a proprietary supply chain intelligence asset spanning 1,600+ brands, 500K+ carriers, 3.6 million facilities, and 11 years of cross-company operational data, enriched with global intelligence layers including customs, sanctions, weather, maritime, and geopolitical data.

An Agent Operating Procedure (AOP) is an executable playbook that defines exactly how an AI agent handles a specific supply chain situation: which data to gather, which intelligence to consult, which action to take, which channels to use, and when to escalate.

AI explainability in supply chain is the ability to provide human-understandable explanations for every AI-generated prediction, recommendation, or action, answering not just 'what did the AI do' but 'why did it do that' and 'what information informed the decision.'

A Decision Trace is a complete, step-by-step record of every action an AI agent takes: what triggered the workflow, what data was gathered, what intelligence informed the decision, what action was taken, what the response was, and what the outcome was.

AI Decision Intelligence is the accumulated library of supply chain decision precedents: what resolution paths worked, for which exception types, on which carrier-lane combinations, in which seasons, captured from millions of real agent actions.

Conversational analytics in supply chain is the ability to query supply chain data using natural language (typed or spoken) and receive instant, data-grounded answers with visualizations, eliminating the need for pre-built dashboards, analyst support, or SQL expertise.

Outcome as a Service is a commercial model in which an enterprise pays for measurable business outcomes delivered by AI agents, rather than software licenses, seats, or modules.

Loft is FourKites' multi-agent orchestration platform where AI agents operate, workflows are built, and every decision is recorded with a full audit trail.