Demand Forecasting Solution: 5-Wk Proof of Concept

Affine Inc

Demand Forecasting using Azure 5-weeks PoC is designed and developed for consumer-driven demand forecasting scenarios across various industries, allowing organizations to accurately predict demand.

Affine’s demand forecasting solution helps to build a highly accurate product demand estimations based on internal factors like product attributes, merchandizing, marketing etc. and external factors such as Competitor influence, geography, seasonality, macroeconomics etc. which assisted stakeholders in short-term, medium-term, and long-term planning for manufacturing & supply chain to meet future demand.

The focus of this PoC is to leverage Azure services to create demand forecasting engine which can be customized to predict demand for individual product at required level of granularity as per business needs enabling Operations teams to plan their production accordingly.

The solution leverages Azure services such as Azure ML Services, Azure Data Factory, Azure Synapse, Azure Data Lake, Databricks, Power BI etc.

Agenda (5-weeks):

  1. Business Understanding - Business Understanding, Hypothesis Creation & Signoff
  2. Analytical Dataset Creation - Data Loading from Internal & External Data Sources, Data Preparation & Analytical Dataset Creation
  3. Analytical Modelling - EDA & Feature Engineering
  4. Demand Forecasting - Model Development (1 Model Applicable for upto 4 to 5 SKU / Product of same category), Model Validation & Hyperparameter Tuning
  5. Generate Results - Generating Demand Forecast which can be published in Power BI Dashboard


  • Improved forecasting accuracy
  • Improved ROI via accurate impact quantification
  • Reduced promotional costs via what-if scenarios
  • Reduced manhours on generating forecasts
  • Reduced planning process

Why Affine

Enabling business focused data science, AI and BI development with deep domain expertise. Affine believes in faster design to faster deployment through key differentiators- Experimentation Focus and Speed to Value.