The future is already here, according to logistics IT pundits. A study of the impact of predictive analytics on logistics and supply chains published by Eyefortransport (eft) argues that artificial intelligence (AI) is rapidly advancing beyond the nice-to-have arena to a necessary asset.

“Predictive analytics is no longer just a helpful bonus feature to have in logistics; it’s a necessity,” its authors proclaimed.

Logility, a supply chain management software provider, also makes a strong case for the deployment of AI in logistics. The speed and complexity of business requires companies to quickly turn information from a multitude of sources into actionable insights, it stated.

According to eft, this sentiment is finding strong resonance among logistics executives. It points to a study from the Council of Supply Chain Management Professionals which found that 93% of shippers surveyed and 98% of 3PLs feel data-driven decision making has become crucial to supply chain activities.

AI in logistics has been associated largely with big multinational firms, above all Amazon, which has been using predictive analytics to anticipate sales trends and position inventory accordingly. However, it can also be employed by small firms. By the same token, predictive analytics can be tailored for anything, from a single warehouse to an entire supply chain, eft finds.

While Amazon famously creates new layers of visibility into seasonal buying patterns and forecasts through predictive analytics, most players are currently looking at a more basic level of improved supply chain visibility to help them avoid late shipments and supply chain disruptions. AI can help identify potential problems, plan for unexpected conditions and optimize processes and route planning.

Four Kites, whose load matching network platform is used by shippers, carriers and logistics firms to collaborate, combines its core tracking algorithm with predictive updates on weather and traffic conditions to update arrival time predictions for shipments. The service is used by a number of Fortune 500 companies. Walmart is working with Four Kites on further development of the retailer’s supply chain visibility and predictive analytics capabilities.

Carriers are turning to predictive analytics for their equipment management. Predictive analytics can detect failure patterns and anomalies and predict failures of components, allowing the companies to schedule timely maintenance work.

At this point much of the early adoption of AI has been in the operational sphere. DB Schenker uses its Decision Support Tool in warehouses to simulate daily scheduling and processes in order to optimize operations.

Maersk has begun to use AI to improve the repositioning of empty containers. The company estimates that this will save it millions of dollars.

UPS sees potential savings between US$100 million and US$200 million through its new AI-supported Network Tools Planning software, which is scheduled to be fully implemented in the US next year.

XPO Logistics is moving to put AI at the fingertips of its clients for forecasting purposes. At the end of May it unveiled a digital dashboard that can be tailored to the unique forecasting models engineered for its customers. According to XPO, this offers a more tactical understanding of how goods are distributed to the point of consumption. The logistics firm uses predictive analytics to realize future efficiencies within the warehouse, providing insights that customers can use to enhance procurement, customer relations and other areas of their business.

“We’ve created a way for customers to access their supply chain data in the form of private business intelligence, while the heavy lifting is done by our analytical models. This is where the value lies for our customers – at the ideal point between a high-level overview and raw data,” said chief information officer Mario Harik.

More companies are bent on utilizing AI. An annual survey of supply chain executives published jointly by JDA Software and KPMG indicates that end-to-end visibility is the top priority, which respondents intend to achieve through the increased use of machine learning and cognitive analytics.

“A truly autonomous supply chain requires predictive end-to-end visibility and these survey results echo our vision to make this a reality for our customers,” said Fred Baumann, group vice president, industry strategies at JDA.

By Ian Putzger

Correspondent | Toronto