Week #2: An Offer We Can't Refuse

1. The proposal from the professors

The team agreed that by the end of this week, the challenge theme must be settled, in order to start studying the state-of-the-art solutions and techniques, besides starting to look for a dataset next week.

Many ideas were discussed, mainly the Product Pricing Strategy problem, in which the team saw potential for an interesting solution that would combine a ML model for price definition with a genetic algorithm that would be able to optimize its hyper parameters, providing the solution that would best improve the sales/profit of a retail company that would use it.

When presented to the team inspirator, professor Tiago Pinto, the team promptly addressed the main challenge that would surface by pursuing this subject: finding a large and rich dataset.

When confronted with this problem, professor Tiago gave a suggestion to the team that would potentially solve any issue there may be with the dataset: the possibility to collaborate with the team’s mentor, Hernani, in the development of his dissertation’s project.

By following this suggestion, on the one hand, the team would be provided with a real supermarket dataset and the possibility to discuss ideas with Hernani, who has already analysed and started working on his solution. On the other hand, Hernani would also benefit from the team’s insights and alternative solution, enriching his dissertation.

As such, the team saw this as a good opportunity to develop a project with great relevance to every retail company, with a real and rich dataset that will, ideally, reflect in a more effective solution.

Despite that, the team and professor Tiago agreed that before any decision was made, a meeting must be arranged with Hernani to discuss further details.

2. The discussion with the team's mentor

Later this week, the team got the opportunity to discuss, with Hernani, his ideas for his dissertation and the respective project, which consisted in Store Layout and Shelf Planning, with the main goal of maximizing the number of sales.

Hernani took this opportunity to explain his perspective on the problem, how he intends to implement a solution and gave the team a bit of context on the data that he will be able to provide, saying that it refers to a supermarket chain in South America, with a significant amount of daily transactions that should make up for a large and rich dataset.

With this in mind and given the benefits identified previously, the team accepted the opportunity to collaborate with Hernani, with further details regarding the problem definition and a dataset analysis coming up on next week's post. Stay tuned!