Transportation teams today are more data-driven than ever before. They demand accurate and timely data to make decisions — because their customers’ expect no less —and they are responsible for any number of KPIs, including OTIF compliance, various operational efficiencies like cycle times, reducing dwell times to improve carrier performance, detention fees to improve margins and better customer service.
For our customers, these targets are achievable by leveraging the FourKites’ platform to their advantage. FourKites ingests, analyzes and acts upon over 1 million data streams that relate to and/or impact freight during its journey. One of the key bits of data that real-time visibility depends upon is the location data for the vehicle that’s transporting the freight itself, usually supplied by electronic logging devices, or ELDs.
But what if you don’t have that data? What if the vehicle doesn’t have an ELD device, you don’t know which ELD is associated with an asset, or the ELD is unable to transmit its location? These aren’t just theoretical questions. For all of the progress we have made bringing visibility to previously dark areas of the supply chain, untrackable loads remain a challenging issue.
With the introduction of our new patented Smart Forecasted Arrival (SFA) capabilities, not only can we now track freight that isn’t transmitting a location, but we can provide highly frequent and highly accurate ETAs for that freight. Specifically, we can provide ETAs on 97% of untracked loads, with 85% accuracy. And we can make correct predictions on late loads more than 90% of the time. It’s worth underscoring: These loads are not just untracked, they are untrackable today.
The patent is certainly merited, as Smart Forecasted Arrival represents a true technical breakthrough — one that only FourKites and its community have the assets and capabilities to pioneer. Here’s why:
Lots and lots of data. FourKites invented the market for real-time transportation visibility, and that first-mover advantage has helped us build the largest global network of supply chain data on the planet. Every year, our platform tracks over 70 billion miles across 275,000 distinct stops, and computes 1 billion ETAs. And these numbers are growing exponentially each year! Every day, we track more than 2.5 Million Shipments across 200+ countries and territories globally, spanning every mode of transportation. This broad, deep and rich dataset is a prerequisite to sophisticated analytics, predictive ETAs and other advanced data science techniques. (More on the latter in a moment).
Data quality. Data quality is a broad topic, encompassing many different dimensions of data, from accuracy to completeness to consistency and much more. FourKites has made significant investments in data quality best practices — including, for example, master data management and data federation protocols — to help ensure that all of the data that our platform ingests is optimized for analysis and other sophisticated product capabilities. We monitor data at an aggregate level, and scrutinize it at a unit level. We also use advanced data science to help monitor data quality. Which brings me to the final piece of the puzzle…
Data science + “Data DNA”. A passion for data and data science is built into FourKites’ DNA. From our inception, company leadership understood that data would be foundational to our vision of a future of automated, interconnected and collaborative global supply chains. Our co-founder and CEO Matt Elenjickal is an engineer by trade, with decades of experience implementing complex enterprise systems. Much of my career and expertise are rooted in data science. Our Chief Product Officer Priya Rajagopalan (recently named a Top 25 Software Product Executive by The Software Report) is an expert in developing cutting-edge products that leverage the latest in data science techniques.
And we are just the tip of the iceberg — the iceberg being the world-class data science team that we have assembled here at FourKites. This is the team that tackled the question I posed earlier — What do you do when there’s no data? — and answered it with groundbreaking new capabilities that use machine learning to fill in the gaps.
That’s the beauty of Smart Forecasted Arrival. It removes blind spots and compensates for poor-quality data, or even a complete lack of tracking data, by intelligently creating that data exactly where it’s needed. SFA would not be possible if not for the industry’s richest and highest-quality data set, aggregated from the largest network of global shippers and their carrier partners; if not for a scalable platform that can process tens of thousands of data providers and trillions of data points; and if not for a corporate-wide commitment to data that is embodied in the industry’s preeminent data science team. It’s a unique and powerful combination, and I look forward to many more groundbreaking capabilities to come.