Price prediction and optimization

Enabled a leading food processing company to efficiently predict supply & demand, and the optimal price of its products

food-processing-banner-small

About the Client

The client is a multinational food company based in the US. The client is a world leader in chicken products enjoyed by millions of customers around the globe. For more than six decades, the client has been distributing its products through retail and food service outlets, domestically, and internationally.

The Business Challenge

The client wanted to effectively predict the demand and supply for its products every month. The client receives orders from commercial & non-commercial food services, retail outlets, and customers both in the US and abroad every month. The client also wanted to come up with the optimal price for each of its food categories by considering these supply and demand drivers.

What Aptus Data Labs Did

We built an ML-based Price Predictive Engine on AWS Sage Maker along with CI/CD Services. Given below is the list of factors considered to resolve the business challenge. The predictive engine considered all these factors, cleaned the existing data, and selected the factors that impact the most to predict the optimal price for each product. 

Factors-analysed-food-processing-casestudy

The Business Impact Aptus Data Labs Made

The Business and Technology Approach

Here is the comparison of what the business was earlier following and what we have proposed.

AI-ML-Based-price-predictive-analysis-casestudy

Work Flow/Architecture/Screenshot

The solution we delivered is currently live on the AWS platform with a CI/CD pipeline in place to ease the process of future model retraining and refining.

aws-solution-flowchart

Tools used

Python Amazon Code Commit
Flask
Amazon ECR
Docker
Amazon Sage Maker
EC2
Lambda
Amazon Code pipeline
API Gateway

The Outcome

The new predictive engine successfully helped the client to predict and estimate the optimal price for all its food categories to be quoted to its customers. The client could also identify the best set of parameters that would impact the price the most. The client was able to overcome the business challenge with an automated solution that saved much time, costs, and resources usually required by traditional manual work.



Related Case Studies

Download Case study

Download Case study