I am currently working on modifying an existing notebook called "Avocado Price Optimization" https://www.gurobi.com/jupyter_models/avocado-price-optimization/ that utilizes machine learning to predict demand and the Gurobi solver to optimize the model. This is for a university research workshop (TID). The modifications I have made include transforming the "price" and "units_sold" variables to a logarithmic scale and calculating the price effect for each region to improve the regression and address the optimization problem.
This project demonstrates the application of data science techniques to solve real-world problems in the agricultural industry. Within the GitHub repository, you can find the modified notebook and accompanying code, as well as a detailed description of the modifications made and the results achieved.
In addition to the "Avocado Price Optimization" project, I am also working on the development of a nonlinear programming model with continuous variables for four products. This new project is based on the work of Alan L. Montgomery "Creating Micro-Marketing Pricing Strategies Using Supermarket Scanner Data" and it aims to solve a nonlinear optimization problem using different solvers available in R.
The objective of this project is to compare the performance of various solvers when dealing with nonlinear optimization problems. By doing so, we can better understand the strengths and weaknesses of each solver and make more informed decisions about which tool to use in various situations. This project is in line with the overall objective of the repository, which is to apply data science techniques to real-world problems.
The GitHub repository will be updated with the code for this new project, along with a detailed explanation of the model, the methods used, and the results achieved.