WFS Deploys Machine Learning Tool to Improve Air Cargo Forecasting and Workforce Planning

Worldwide Flight Services (WFS), a SATS company, has developed a new machine learning-based forecasting tool designed to improve the accuracy of air cargo volume predictions and better align workforce resources across its global network.

The digital platform, known as the Performance Management Platform – Machine Learning Forecast (PMP MLF), uses algorithms trained on 10 years of operational data, including more than three million air waybills, historical flight and truck movements, seasonality, holidays, and cargo types. The system generates daily forecasts of cargo tonnage, ULDs, and piece counts by flight, truck, customer, transport mode, and warehouse location.

WFS says the tool addresses long-standing forecasting challenges in air cargo, where volatile volumes and reliance on manual estimates or historical averages have often resulted in staffing gaps of 10 to 15 percent. By providing more accurate forward-looking data, PMP MLF enables stations to plan labor and resources in advance, reducing reactive operations, service inconsistencies, and inefficiencies.

The platform currently supports forecasts for nearly 10,000 flights and more than 6,000 truck movements per week across 75 warehouses in 13 countries. Forecasts feed directly into station-level planning tools, allowing teams to identify potential volume surges early and adjust staffing proactively between teams or locations.

According to WFS, performance data shows the system achieves forecast accuracy levels between 92 and 98 percent, including during periods of irregular demand. The company says this has helped reduce service level agreement breaches, avoid unnecessary overtime, and limit idle labor time.

Phase two of the platform was rolled out in summer 2025 and introduced enhanced dashboards, improved visual analytics, tighter integration with workforce management and rostering tools, and customer-level forecasting to support collaborative planning of volume peaks.

“For many years, cargo handlers have relied on manual scheduling, spreadsheets, or basic rolling averages for forecasting,” said Jimi Daniel Hansen, senior vice president of operational excellence at WFS. “By leveraging machine learning within a complex operational network, our goal was to replace reactive guesswork with data-driven clarity to optimize workforce allocation, enhance service levels, and reduce operational waste across our global air cargo network.”

He added that the improved forecasting capability translates into fewer delays related to staffing issues, greater service consistency, and more transparent, data-backed capacity planning for customers.