USA | The Massachusetts Institute of Technology (MIT) Centre for Transportation & Logistics and Mecalux have developed an artificial intelligence-based simulator capable of optimising inventory distribution across different warehouses within the same logistics network.
The platform, called Genetic Evaluation & Simulation for Inventory Strategy (GENESIS), uses advanced machine learning models to analyse thousands of possible scenarios and determine the optimal stock level at each warehouse and when replenishment should occur.
The AI-based simulator accounts for variables such as forecast demand in each region, transport costs, and the operational capacity of each warehouse to test various inventory replenishment policies without affecting real-world operations.
“The genetic algorithm enables multiple simulations to be run using different parameters until the most efficient logistics strategy is identified. Companies can compare scenarios and select the one that best fits their operations,” says Dr Matthias Winkenbach, Director of Research at the MIT Centre for Transportation & Logistics and the Intelligent Logistics Systems Lab.
Once data and variables are entered into the system, GENESIS generates the optimal solution along with advanced statistical dashboards. Users can analyse indicators such as consumption patterns, regions with high demand variability, SKUs with a higher risk of stockouts, and warehouses experiencing supply issues.
Redistribute before purchasing
One of the system’s key features is its ability to rebalance inventory across warehouses. Instead of automatically placing new orders with suppliers, the tool analyses whether it is more efficient to transfer products from another facility within the network where excess inventory is available. In this way, companies can reduce costs and better utilise existing stock.
The system also recommends how to organise transport. For example, it suggests whether shipments should be consolidated to optimise truckloads or whether specific orders should be fulfilled from a particular location to reduce delivery times and costs.
“The real challenge wasn’t finding the right algorithm; it was making it fast enough to be practical. We developed GENESIS from the ground up to evaluate thousands of scenarios simultaneously rather than sequentially. What used to take days now takes minutes, which means companies can use it for real tactical planning, not just theoretical analysis,” saidRodrigo Hermosilla, Research Engineer at the MIT Intelligent Logistics Systems Lab.
Upcoming AI applications
The AI-powered simulator is one of the first tangible results of the joint initiative between Mecalux and MIT CTL. The collaboration has now entered a new phase focused on expanding the application of AI to other logistics processes, such as internal replenishment, digital twins in high-density automated storage systems, and slotting optimisation.
More tech news here
