The FourKites Knowledge Center

Carrier capacity planning is the forecasting and management of transportation capacity needs by lane, mode, and time period, ensuring that sufficient carrier resources are available to meet anticipated shipment volumes without over-commitment or spot market exposure.

Total Cost of Logistics (TCL) is the comprehensive measurement of all costs associated with moving goods through the supply chain, including transportation (freight rates, fuel surcharges, accessorials), warehousing (storage, handling, labor), inventory carrying costs, administration, and the cost of service failures (penalties, expedited shipments, lost sales).

Cross-Company Intelligence is the layer of supply chain insight that emerges only when operational data from multiple enterprises is aggregated: carrier behavior patterns, facility benchmarks, lane performance distributions, and exception resolution precedents that no single company could observe alone.

Sophie is FourKites' AI Developer Agent that translates natural-language descriptions of supply chain processes into production-ready Agent Operating Procedures (AOPs), connected to the Graph and Digital Twins, deployable in days.

AI guardrails are the boundaries, constraints, and safety mechanisms that limit what an AI agent can do in production, preventing actions that could cause financial, operational, or reputational harm, including confidence thresholds, action limits, approval gates, and rollback capabilities.

When deploying AI agents in supply chain, 80% of every workflow should leverage proven patterns from the network (pre-built AOPs validated across Fortune 500 deployments) and 20% should be customized to the customer's specific business rules, escalation paths, and integration requirements. This framework explains why this ratio matters and how to apply it.

The distinction between supply chain AI powered by a multi-enterprise network (cross-company intelligence from hundreds of shippers) versus AI powered by a single customer's data (limited to one company's operational history).

Air freight visibility is the tracking of cargo shipments transported by aircraft, from airport of origin through potential hub transfers to airport of destination, including flight schedule adherence, customs clearance, and ground handling milestones.

Gen UI (Generative UI) is a capability within FourSight AI that dynamically generates custom, persona-specific supply chain interfaces without manual configuration or dashboard building.

AI confidence scoring is the assignment of a numerical score to each AI-generated prediction, recommendation, or action, indicating the system's estimated probability that the output is correct, enabling graduated response strategies based on certainty level.

For 25 years, enterprise software followed one model: build tools, sell subscriptions, charge per seat. The buyer got access to software. What they did with it was their problem. AI is inverting this model. The new category, Service-as-Software, sells completed work: exceptions resolved, appointments scheduled, documents processed. The buyer pays for outcomes delivered, not software accessed. This framework examines the structural differences, the economic implications, and why the transition is happening now.

Lane performance is the measurement of transportation reliability, cost, and service quality on a specific origin-destination pair (lane), encompassing on-time rates, transit time consistency, exception frequency, carrier mix, and cost per load.

Most enterprise AI ROI claims are based on projected value, not measured outcomes. 'We estimate $2M in annual savings' is not the same as 'Tracy resolved 12,400 exceptions last month, eliminating 4,100 manual carrier calls, saving 2,050 labor hours at $45/hour loaded cost, for a measured value of $92,250 this month.' This framework provides the measurement methodology for supply chain AI ROI that separates actual outcomes from estimated potential.

Most AI competitive advantages are temporary. Better models are published monthly. Better prompts are shared on Twitter. Better UIs are built in weeks. But data moats, specifically network intelligence that compounds with every customer, every shipment, and every agent action, create structural advantages that widen over time. This framework examines why network effects in AI data are the most defensible moat in enterprise software and how FourKites built one.

Demurrage is the fee charged when a container remains at a port or terminal beyond the allotted free time period (typically 3-5 days at US ports), calculated per container per day at rates that escalate with duration.