AI powered sustainable inventory & policy monitoring 2

Welcome to this presentation about artificial intelligence, inventory management, aligned with sustainable operations and monitoring policies. My name is Antonio Pesqueira. I'm affiliated with ISCTE, University of Lisbon, and also with an organization called DCOPI, that is the Dynamic Capabilities Operational Institute that mainly does research and also activities around inventory and supply chain management strategy.

I am an expert in life sciences, healthcare operations, and supply chain management. My specialization goes in terms of implementing advanced digital technologies like AI and blockchain.

To make it simple, there are 10 different ways where AI can really transform inventory management. One can be the different ways AI can really have more efficiency regarding risk management, the ways we can work more with the sensors, the way that we can have a more capable demand forecasting, the way we can have as well a smarter warehouse management with AI robots, data mining, order optimization, supplier relationship management, and so on and so on. As we look at the slide, we can see as well one specific area that is extremely important is the green optimization and the Internet of Things, or the IoT, condition monitoring.

There basically overall, is the optimization can be connected with different real-world data points, where we have edge-based IoT sensors that can send temperature and vibration readings to real-time optimization solvers, and these solvers can make prioritization in terms of the first-in, first-out picks. Those first-in, first-out picks they keep the products intact, and they might avoid extra costs from reworking. Or when we have multi-objective routing mechanisms to try to minimize the distance, emissions, and the risk of a product being the first to expire and then to be removed from the inventory lines. These mechanisms from AI through machine learning, for example, can also change the carrier used as the order fulfillment starts queuing different changes during the day. And imagine the huge capacity that this can help us in terms of having calculations around the safety stocks when we can start having life quality scores to adjust different inventory levels. We can have those sensors, for example, within pharmaceuticals, to start having better predictions around the cold-chain stability when the buffers start getting smaller. When the sensors start to signal different problems that might suggest that products will soon be out of date, then the buffer starts to get bigger. Automatically, those alerts start to be more efficient when we start working with different production sites, when we start acting more effectively with the different distribution centers as well.