Recommendation Engine uses a combination of historical data such as average dwell at origin facility and real-time transit conditions, including traffic and weather, to make proactive recommendations in real-time to mitigate potential problems before they occur.
FourKites gives a “Departure from Origin” recommendation for the shipment to make an on-time delivery. The company says that early users have been able to improve load planning accuracy, understand delivery risks before the load is even picked up, and reduce fees related to late reschedules, detention, and missed appointment times.
“As supply chains grow in complexity, companies can no longer rely on approaches that worked in the past to plan their shipments,” said Priya Rajagopalan, chief product officer at FourKites. “The industry must evolve and become more proactive, harnessing data and intelligence that can provide actionable insights, and anticipate and mitigate problems to ensure on-time delivery of a shipment before a load has even left the dock.”
Traditionally, the use of analytics in supply chain could identify historical patterns in data, but failed to detect and prevent problems in advance, the company says. Now shippers and 3PLs can identify potentially late loads before they are even picked up to proactively work to reschedule the load or implement a recovery plan and notify their customers of possible delays.
The system can also provide real-time alerts when a shipment deviates from the most optimal route or when new risks appear. The introduction of FourKites’ Recommendation Engine follows the recent launch of a DynamicETA machine learning algorithm that predicts freight arrival times with greater accuracy in narrow time windows.
FourKites says it has a customer network that includes more than 250 of the world’s top shippers, including 18 of the top-20 food and beverage companies and 4 of the top-5 CPG companies. FourKites tracks more than 500,000 shipments every day.
