Amazon Supply Chain and Logistics

AWS Supply Chain announces new upstream capabilities

We announced four new capabilities for AWS Supply Chain at re:Invent 2023 to support upstream supply chain processes.

  1. AWS Supply Chain Supply Planning to help you plan, position, and replenish components and finished goods to reduce inventory costs and respond more quickly to demand variations and supply disruptions.
  2. AWS Supply Chain N-Tier Visibility to streamline communication between you and your suppliers, improving your ability to more accurately respond to supply plans as well as manage demand or supply changes during the execution window. With this capability, you can securely collaborate with your trading partners in just a few clicks.
  3. AWS Supply Chain Sustainability to provide a central repository allowing you to request, collect, and audit sustainability data.
  4. Amazon Q in AWS Supply Chain, which uses generative artificial intelligence (AI) to provide supply chain professionals with a conversational experience to aid in analysis and decision-making.

These new capabilities, which will be available in 2024, expand existing data lake, demand planning, and machine learning (ML)-powered insights. Today, customers are using AWS Supply Chain to improve inventory visibility, help prevent stock-outs, and reduce overstock events that can increase carrying costs. AWS Supply Chain makes this possible by aggregating relevant customer data from disparate Enterprise Resource Planning (ERP) systems into a unified, canonical data model and creating a supply chain data lake (SCDL). This unified SCDL enables Insights, an AWS Supply Chain capability that provides a comprehensive view of the supply chain to improve inventory visibility, and provides machine learning-powered recommendations to help mitigate inventory and lead-time risks. The Demand Planning capability combines Amazon’s deep supply chain expertise with ML to analyze historical sales data and real-time data to create forecasts and continually adjust models to improve accuracy.

Upstream supply chain challenges

The upstream part of the supply chain includes sourcing and movement of raw materials and components from suppliers and trading partners. Supply chain leaders have shared that they continuously face the challenge of coordinating with many different layers of trading partners such as suppliers, manufacturers, distributors, and retailers. Each trading partner often has their own data management systems that require expensive customization and long development cycles or manual work-arounds to integrate. As a result, supply planners spend a lot of time reconciling forecasts, order confirmations, shipment quantities, and more. Siloed data—coupled with demand variations, supply disruptions, and vendor lead-time uncertainty—makes it difficult for companies to accurately determine and position inventory to meet the customer demand

Manufacturers face additional complexities associated with raw materials and components, which are prone to variable pricing and availability. Furthermore, trading partner responses to customer requests for data vary in quality, frequency, timeliness, and structure, and are not always systematically tracked or audited.

Managing compliance artifacts like carbon emissions and hazardous material disclosures at scale is similarly challenging, and has traditionally been done via email, fax, and messaging apps, without formal tracking and auditing mechanisms. As a result, many organizations struggle to ensure appropriate quantities of goods are in the right place at the right time to efficiently meet demand or to meet increasingly stringent regulatory requirements.

With these new AWS Supply Chain capabilities, you can improve your upstream supply chain processes, improve availability or in-stock rate of materials, communicate with suppliers to confirm supply plans and obtain commitments, and get accurate data about key environmental factors. Let’s review these capabilities in more detail

AWS Supply Chain Supply Planning

Supply Planning is based on Amazon’s expertise in developing sophisticated supply planning models for its own operations. Supply Planning creates sophisticated supply planning models that can accurately determine the right levels of inventory needed across facilities. Supply plans are generated using the demand forecasts created by AWS Supply Chain Demand Planning, combined with product, facility, bill of materials (BOM), inventory, and other customer information from the SCDL. This auto-integration of data improves information quality and reduces the risk of error by replacing manual efforts to consolidate various reports containing forecasts, order confirmations, and vendor lead times.

Supply Planning dynamically calculates inventory targets while accounting for demand variability, vendor lead times, and ordering frequency to recommend purchase order creation or inventory transfers. This will help you determine the optimal number of units to order or move, when to place orders or transfers, and where to position inventory.

The following screenshots shows the Supply Planning dashboard that summarizes on-hand and on-order stock, purchase order status, and operational metrics. You can examine each of the categories in more detail and take the appropriate steps to track, follow up, or view more information.

Supply Planning Dashboard

Supply Planning Insights

AWS Supply Chain N-Tier Visibility

N-Tier Visibility extends your visibility and insights beyond your organization to multiple external trading partners. You can invite and onboard trading partners in just a few clicks and view all partners across your network as shown in the following screenshots.

N-Tier Partner List

N-Tier Partner Map

This connectivity also enables trading partners to automate communication and improve their own forecasts. For example, you can share purchase orders and supply forecasts with your trading partners, and then track the status of those purchase orders or changing inventory levels, from within AWS Supply Chan N-Tier Visibility. The updated supply plans and purchase orders are exported to Amazon Simple Storage Service (S3), so you can integrate them with your enterprise resource planning (ERP) systems.

The built-in chat and messaging capabilities make collaboration even easier across the entire supply chain. For example, if a component shipment is delayed, an inventory manager can message a supplier to identify a work-around within the AWS Supply Chain application. Improved internal collaboration and information sharing with suppliers and manufacturers enhances your ability to detect sourcing risks and component shortages, empowering you to quickly mitigate disruptions.

AWS Supply Chain Sustainability

Sustainability creates a more secure and efficient way for you to obtain mandatory documents and datasets from your supplier network. You can request, collect, and export artifacts such as product life cycle assessments, certificates on product safety, or reports on hazardous substances used at any point in the supply chain.

You can also upload your own data collection form for your suppliers to document any sustainability issue and send data requests to multiple tiers of trading partners, track responses, send reminders to absentees, and provide a central repository to store and view responses. This is shown in the following screenshot.

Sustainability Info Request

In cases of supplier carbon emission data responses, you can also visualize the aggregated data as shown in the following screenshot. These capabilities will simplify environmental and social governance (ESG) compliance reporting and eliminate additional manual processes and risk of error.

Sustainability Emissions Dashboard

Amazon Q in AWS Supply Chain

Finally, AWS Supply Chain will have a generative AI assistant powered by Amazon Bedrock that provides a natural language interface so you can query data within the SCDL, and receive intelligent answers to “what?” “why?” and “what if?” questions. Questions may be asked via a conversational prompt within the application. Amazon Q can also visualize outcomes of complex scenarios and the tradeoffs between different supply chain decisions. Example question-answer pairs can look like the following:

Q1: How many brake pads are we ordering?

A1: We are ordering 1500 units in brake pad category across 5 products.

Q2: Why did the brake pads order quantity increase vs last month?

A2: Order quantity increased by 50% due to higher lead time from vendor AA of 56 days this month vs 28 days last month.

Q3: How will the order change if the supplier reduces lead times by 2 weeks?

A3: If lead time for brake pads from vendor AA reduces by 2 weeks to 42 days, the order quantity will reduce by 20%.

The response is shown in the following screenshot.

Amazon Q in AWS Supply Chain conversation

This is a hypothetical example to demonstrate the depth and breadth of Amazon Q in AWS Supply Chain. The speed and accuracy of Supply Chain Q’s responses will help you quickly gain insights, understand causality, and identify potential options to improve your supply chain performance.


AWS re:Invent is an exciting event for our customers, partners, and for AWS because we can meet and learn from each other and share our newest innovations. This year we announced four new capabilities that will further improve supply chain planning and visibility across multiple tiers of our customers’ supply chains and address several key challenges. Supply Planning, N-Tier Visibility, Sustainability, and Amazon Q in AWS Supply Chain will be available in 2024.

Visit AWS Supply Chain to learn more and get started. For a self-paced technical overview, visit AWS Workshops. Please also visit our re:Invent page to access re:Invent session recordings and additional useful resources.

Diego Pantoja-Navajas

Diego Pantoja-Navajas

Diego Pantoja-Navajas is the Vice President of AWS Supply Chain and is responsible for the vision and execution of business applications. He and his team have reimagined how supply chains can operate and are focused on bringing the world’s first continuously improving supply chain system of record to the market. He is passionate about his customers’ success and using SaaS, cloud, and AI/ML technologies to build highly usable and intelligent B2B enterprise software solutions to solve business problems related to supply chains, e-commerce, and fulfillment. Diego is an honor graduate of the Georgia Institute of Technology and has continued his training, completing executive education courses in Artificial Intelligence & Machine Learning at MIT. He has also participated in multiple leadership courses in partnership with IESE Business School and the University of Michigan, Ross Business School. He lives with his family in South Florida and is always happy to learn more about innovative products or solutions that will continue driving business success for his customers.