About Generating Demand Forecasts at Scale Using Oracle Cloud Infrastructure

Demand forecasting is the process of leveraging historical data and other analytical information to build models that help predict future estimates of customer demand for specific products over a specific period. Demand forecasting helps shape product road map, determine inventory production and inventory allocation, to name a few. According to McKinsey, a 10% to 20% improvement in supply chain forecasting accuracy is likely to produce a 5% reduction in inventory costs and a 2% to 3% increase in revenues. In a world where margins are increasingly narrow and critical, this percentage can be make or break. Traditional supply chain forecasting tools have failed to deliver the desired results, limiting success of retailers and manufacturers.

This architecture enables business users in the retail industry to overcome the technical limitations of legacy data analytics solutions that undermine forecasting accuracy. Instead, perform full forecasts on atomic-level data within tight service windows.

Before You Begin

Before you begin, ensure you have completed the following prerequisites:

• Create an object store bucket.
• Download the Store Item Demand Forecasting Challenge dataset to your local machine. Unzip the train.csv.zip file, and upload the decompressed .csv file to your object store bucket.
• Create and launch an OCI Data Science notebook instance. For instructions, see Use Interactive OCI Data Science Notebook with OCI Data Flow.

Architecture

This architecture builds a fine-grained demand forecast at the store-item level. Use this architecture to build a demand forecasting solution that leverages the power of Oracle Cloud Infrastructure (OCI).

This architecture leverages the following OCI services:

• OCI Object Storage

  An internet-scale, high-performance storage platform that offers reliable and cost-efficient data durability. The Object Storage service can store an unlimited amount of unstuctured data of any content type ranging from analytical data to rich content such as images or video. Object Storage safely and securely enables you to store or retrieve data directly from the internet, or from within the cloud platform.

• OCI Data Flow

  A fully managed Apache Spark service that performs processing tasks on extremely large datasets - without infrastructure to deploy or manage. Developers can also use Spark Streaming to perform cloud ETL on their continuously produced streaming data. This enables rapid application delivery because developers can focus on app development, not infrastructure management. Data Flow now pre-includes Delta Lake libraries (1.2 & 2.0) with Spark. Data Flow studio feature unlocks interactive Spark queries at scale (on TB to PB size datasets) from a managed Jupyter environment powered by OCI Data Science.

• OCI Data Science

  A fully managed platform for teams of data scientists to build, train, deploy, and manage machine learning models using Python and open-source tools. Use a JupyterLab-based environment to experiment and develop models. Scale up model training with NVIDIA GPUs and distributed training. Take models into production and keep them healthy with MLOps capabilities, such as automated pipelines, model deployments, and model monitoring.

About Required Products and Roles

This solution requires the following products and roles:

• Oracle Cloud Infrastructure Object Storage
• Oracle Cloud Infrastructure Data Flow
• Oracle Cloud Infrastructure Data Science

These are the roles needed for each product.

Product Name: Role	Required to...
Oracle Cloud Infrastructure Object Storage: admin	Create object storage buckets.
Oracle Cloud Infrastructure Data Flow: dataflow-admins	Manage and run applications.
Oracle Cloud Infrastructure Data Science: Administrators	Create data science notebook.

See Oracle Products, Solutions, and Services to get what you need.