At their global event, the Seattle-based team is lifting the veil on the AWS supply chain. Designed to aggregate supply chain data, it relies on machine learning to anticipate the risks of interrupting the flow.
Amazon Cloud used its global AWS Re:Invent event, which runs from November 28 to December 2, to showcase the AWS supply chain. This will no doubt be one of the top announcements of this 2022 edition as supply chains have been shaken like never before in the wake of the Covid-19 crisis. The cloud-based application, currently available in preview, consolidates supply chain management data before applying predictive models to predict disruption risks. Faced with historic supply chain software publishers led by Manhattan Associates, Blue Yonder or Körber, Amazon will have to take its place. The information also comes two weeks after Microsoft launched its own supply chain management solution.
Faced with competition, Amazon is betting on artificial intelligence. Upstream, AWS Supply Chain activates learning models to analyze, extract, and transform useful data from third-party sources: integrated management software packages, EDI, other supply chain systems… Problem? Combine this information into a single data model. Further down the chain, machine learning also helps refine supplier lead time predictions. “They will then allow you to optimize the static assumptions that are integrated into your planning models to mitigate the risk of out-of-stock or over-stock,” says one AWS employee. The underlying technology is the same as Amazon.com.
“In order to quickly reach consensus on the actions to be implemented, the tool combines collaboration, messaging and chat features.”
Still relying on artificial intelligence, AWS Supply Chain continues to evaluate various rebalancing options. Recommendations that will be ranked according to the percentage of risk they can solve. At the same time, the tool allows you to analyze their impact on other distribution centers in the network. Depending on the decisions made at the end, the base model is enriched over time. “In order to quickly reach consensus on the actions that need to be implemented, this tool also integrates collaboration, messaging, and chat functionality,” AWS clarifies.
The demand planning block is designed to adjust forecasts regarding market conditions. This is where machine learning comes into play again. To refine his forecasts, he uses both sales history and current trading data feeds, in particular the latest open orders.
Preview available in France
Finally, AWS Supply Chain creates an interactive map that updates in real time (see image above). Purpose: To provide a global view of the state of the supply chain. This allows you to visualize at a glance the amount of inventory in transit and by location, as well as potential out-of-stock risks.
Interestingly, AWS Supply Chain Preview is available in the US, US Cloud US East and US West regions, and Europe in the France region.