Semiconductor Supply Chain Resiliency with AWS

Introduction

AWS can help Semiconductor companies with their value chain transformation by migrating the entire PLM-CAD-ERP workflow to the cloud while mitigating resource and performance risks. Data ingested from SAP into AWS-powered data lakes and Amazon SageMaker for machine learning is used for predictive analytics and simulation of various scenarios. For example, to address supply chain roadblocks or demand spikes. Predictions can be visualized via Amazon QuickSight for engineering & business intelligence.

Engineering teams can use Engineering Design Applications & Desktop solutions on AWS to provision tools in secure, standardized design flows, using compliant methodologies. Administrators of this cloud infrastructure also have the ability to centralize and manage the AWS infrastructure in a cost-effective way. Using these solutions helps businesses facilitate remote workforce collaboration, resulting in improved productivity, while also helping them meet security and performance requirements.

Proven Methods

AWS has proven methods to help you establish your data lake and benefit from it. We can help you set it up and get it running in a remarkably short period of time. A data lake on AWS provides rapid time to value for you to get more from your existing systems and investments today. To start building your data lake on AWS, ingest data into Amazon S3 through 11 different AWS Services.

In the next step, you will manage security with  AWS Services, such as AWS Identity and Access Management (IAM) and AWS Key Management Service (KMS). You will then need to catalog the data using the open data format of your choice (like ORC and Parquet) and tag the data in a central, searchable catalog. Now that the data is in S3, you can use AWS Glue to start analyzing your data and putting it to use in minutes. You can also apply traditional SQL analytics by leveraging Amazon Redshift or Amazon Athena, or use Amazon QuickSight for business intelligence. Let’s look at an example reference architecture below:

Figure 1: Reference Architecture – SAP with Engineering & Design Flow

This reference architecture shows how an on-premises data center connects with an AWS cloud Engineering & Design VPC. It shows how on-premises manufacturing execution systems (MES) logs and ERP applications are connected to the cloud using Amazon API Gateway, and how the factory network is connected via AWS IoT Greengrass services. Once the data is ingested and processed, both structure and unstructured, AWS Athena can be used for data preparation tasks like building views of the workload report data and finally with AWS QuickSight, you can query and build visualizations or dashboards to discover insights from your business and engineering metrics.

The reference architecture shows the process and AWS Services that bring data to actionable insights, and provide a decision support platform to improve semiconductor supply chain resiliency. Below is a more generic and high-level view of the AWS services stack built on top of external data, ERP systems and partner eco-system:

Figure 2: Supply Chain Framework

Conclusion

Running a federated ERP system on AWS, a single source of information, in combination with analytics and AWS Data Lake architectures on AWS can provide up to date, real-time insights across all points in the supply chain and improve the ability to address supply and demand problems, better than what can be achieved with legacy on premises systems.

Leveraging AWS services and solutions, supply chain resiliency for the semiconductor eco-system can be well architected and managed. For more information about running your workflows on AWS: AWS Semiconductor and Electronics Resources