FourKites® Uses Enhanced Machine Learning Model to Predict Shipment Arrival Times With Greater Accuracy in Narrower Time Windows

Share Article

Early Adopters Achieve 90+ Percent Accuracy Predicting Arrivals Within Hours

News Image

FourKites®, the leader in predictive supply chain visibility, today announced a new DynamicETA (SM) algorithm that predicts freight times with greater accuracy in narrow time windows. The enhanced algorithm uses more than 150 data points associated with a single load, and learns continuously from millions of monthly shipments, providing shippers and carriers with more precise arrival times than ever before possible.

Early adopters of FourKites’ new feature have achieved unprecedented results. For example, one top food and grocery wholesaler with greater than 30,000 weekly deliveries achieved 91 percent accuracy in predicting arrival times within a one-hour window, and 97 percent accuracy predicting arrivals within six hours. This allowed the user to confidently communicate with customers and coordinate delivery operations and staffing plans to improve efficiency.

This compares to traditional accuracy rates of roughly 45 percent for day-of arrivals, which are based on top-down estimates using limited data – typically standard drive time of the projected route, adjusted by the remaining hours of service. This low level of precision is insufficient when it comes to meeting modern supply chain requirements, from increasingly stringent customer demands to OTIF and other penalties imposed for late deliveries.

“Real-time data and predictive intelligence are critical to the future of supply chains,” said Mathew Elenjickal, founder and CEO of FourKites. “The industry must evolve from occasionally accurate day-of arrival times to consistently accurate hour-of arrival times. Real-time visibility was the first step, and now we’re demonstrating the power of AI and large data sets to dramatically improve supply chain efficiency.”

FourKites’ DynamicETA is based on a system of best-in-class data models working together to generate the most accurate predictions. Unlike other ETA algorithms, which often track just a small number of attributes, DynamicETA analyzes the impact of 150 different load attributes on arrival time – including shipper, carrier, lane, rest patterns, load, traffic and weather. Each model is customer-centric, generated through intensive and ongoing analysis of all loads and associated attributes. The system is then automatically re-trained every month to utilize recent history and account for seasonal and annual changes in behavior.

“Data science is an important foundation of our innovation process,” said Arun Chandrasekaran, CTO at FourKites. “We’re using data science in two important ways: 1) To improve predictive intelligence for customers, leveraging terabytes of aggregated data to generate valuable new features; and 2) To improve data quality by applying descriptive and inferential statistics that provide accurate and comprehensive visibility into B2B shipments.”

Today’s announcement comes just a few weeks after FourKites announced $50 million in new funding to fuel growth of its real-time visibility network, bringing total capital raised to $100.5 million. Since launching five years ago, FourKites has pioneered the industry shift to real-time visibility, enabling shippers to improve on-time delivery and optimize their supply chain based on actionable data and predictive intelligence. FourKites’ network now includes more than 250 of the world’s top shippers, including AB InBev, Conagra Brands, Kraft Heinz, Nestlé, Perdue Foods, Smithfield Foods, Unilever, Walmart Canada and many others.

FourKites and DynamicETA are service marks of FourKites, Inc.

About FourKites
FourKites is the largest predictive supply chain visibility platform, delivering real-time visibility and predictive analytics for the broadest network of Global 1000 companies and third-party logistics firms. Using a proprietary algorithm to calculate shipment arrival times, FourKites enables customers to lower operating costs, improve on-time performance and strengthen end-customer relationships. With a network of more than four million GPS/ELD devices, FourKites covers all modes, including truckload, LTL, ocean, rail, intermodal, last mile and parcel. The platform is optimized for mobile and equipped with market-leading end-to-end security.

To learn more, visit

Share article on social media or email:

View article via:

Pdf Print

Contact Author

Jessica Presto
Big Valley Marketing for FourKites
+1 (310) 279-6812
Email >
Visit website