OVERVIEW
The objective of this project was to predict the weekly sales for a Retail corporation by store type, store location based on past data and future economic parameters like CPI, fuel price, unemployment. This helped the client to improve cash flow and manage the Supplydemand for the products sold. The future economic forecast data is collected from external sites and databases using Robotic Process Automation (RPA).
Key Benefits (Minsky)
User-Friendly, cloud-based AI platform
No coding skills are required for results or predictions.
Provides you a list of % dependency features that can be used to optimize your business
Ability to fine-tune or optimize the models by trying different algorithms/prediction attributes
Easy integration with other third-party solutions such as TABLEAU for data visualization
Results
Improved Supply-Demand forecast accuracy
Increase in ROI
Minimizes obsolete inventories
Budgets business expenses.
Justifies hiring decisions.
Executes strategic planning.
Improves production scheduling.
Manage cash flow and credit
Executive Summary
The client is a US based Multinational Retail Corporation that has a chain of hypermarkets, discount department stores and grocery stores. The objective was to predict the weekly sales by store type, store location based on economic parameters like CPI, fuel price, unemployment that helps them to make more detailed analysis and improve their sales forecast accuracy. After a detailed evaluation of their operations, AI Labs (www.ailabs.inc) used its proprietary Minsky AI Engine to optimize the models by using a combination of AI algorithms and prediction attributes. In this case, Minsky used historical data for model creation and created weekly sales forecasts by using real time economic data along with store specific information. This solution was optimized and implemented in less than a week.
Typical Sales Forecast Challenges:
Manufacturing the wrong product mix.
Wastage of perishable items.
Increase in obsolete inventories.
Uncertain forecasts lead to wrong hiring decisions.
Excessive time for generation of forecasts
Unable to maximize profit margins /ROI
Inability to manage fluctuations between demand and supply.
Solution
After thoroughly evaluating the client‘s challenges, we used Minsky to accurately model historical data to generate weekly Sales Forecasts. This process includes using weekly sales historical data of past years by store locations and store types. The data also includes weekly economic data, holidays and Sales Promotions. Economic data included parameters like CPI, Fuel price, Unemployment rate in the region of each store which helped us to make more detail analysis. Once the AI Models were generated by Minsky for the selected Algorithms, future economic data along with Store Promotions and holidays were used by Minsky to generate future sales Forecasts. The future economic forecast data is collected using Robotic Process Automation (RPA) from 3rd party sites. Prediction data from Minsky was also integrated with 3rd Party application like Tableau. In addition, RPA was also used to email weekly Sales Forecast generated by Minsky to desired personnel.