AI to future-proof the global supply chain
Efficient supply chains are the backbone of the world economy. But as the COVID-19 pandemic made clear, global supply chains can easily be disrupted. That’s why, as we slowly segue into a post-pandemic world, it is essential that we work to better protect them against future disruptions. “Global economic growth has put a significant strain on our very limited logistics infrastructure,” says Enrique Onieva, a professor in Computing and Intelligent Systems at the University of Deusto in Spain. “The efficient use of this infrastructure and of available transportation resources is a critically important goal.” This is where the project LOGISTAR (Enhanced data management techniques for real time logistics planning and scheduling) comes in. “By taking advantage of the increasingly real-time data gathered from the interconnected environment, the LOGISTAR project aims to clear the way towards the effective planning of transport operations within the supply chain,” adds Onieva, who serves as the project coordinator.
An end-to-end architecture
The main outcome of the project is an end-to-end architecture that can automatically capture and harmonise data, send the corresponding messages to the modules in charge of executing different algorithms, and gather results to be displayed to stakeholders. “This solution takes real-time available data and feeds it to AI-based algorithms,” explains Onieva. “These algorithms are then used to run a number of services, each of which is geared towards optimising supply chain operations.” For example, one service uses precise estimated time of arrival prediction and incident detection to optimise warehouse operations. “By helping warehouses use their available resources more efficiently, this service reduces wait times and the supply chain bottlenecks such delays cause,” adds Onieva. Another service improves how freight is routed and load capacity optimised. “By taking advantage of different modes of transportation, such as trucks, trains and ships, we’re able to both optimise the use of all available infrastructure while also reducing the overall costs of logistics transport,” he notes. On this note, the project also developed a tool for horizontal collaborative planning. “By helping different supply chain stakeholders share available resources, we can reduce the number of kilometres travelled by empty trucks, which in turn reduces greenhouse gas emissions,” says Onieva.
Practical answers to real problems
Onieva says COVID-19 didn’t just disrupt supply chains, it also impacted research projects like LOGISTAR. “The pandemic struck just as we were about to start our test activities, forcing us to conduct everything remotely,” he explains. Despite this unforeseen challenge, the project succeeded at providing practical answers to real supply chain problems. “Our success is a direct result of the commitment of all the people involved in this project,” concludes Onieva. “We may have been a large team coming from different backgrounds and sectors, but we all shared the same objective of future-proofing our global supply chains.” Although the project is now finished, Onieva and some of the project’s other partners are working to improve the technology readiness level of LOGISTAR services. The end goal is to advance the technology towards commercialisation.
Keywords
LOGISTAR, supply chain, automation, artificial intelligence, AI, data, logistics, infrastructure, transportation, algorithm, warehouse
