Awelé or Mancala game - Lycée Val de Durance (Pertuis) Colegiul National Emil Racovita (Cluj)

At first, we wanted to understand the game better, therefore we tried all sorts of online apps. After a few days we saw how many different versions there are, some are solvable while others are not yet completed. All of the teams who contributed to this research decided to use an app as a first reference. We analysed the algorithm of the AI which the app was using and tried to optimise it. In order to do this we needed to find some basic strategies to create our starting point.
So, we competed with our teammates and tried to solve the puzzles the app had to offer, which helped us make the best moves in certain situations and develop some techniques.
While the Romanian teams used the app to construct strategies and analyse the software trained to win Mancala, the French team spent a good amount of time calculating specific positions. They considered that because the value of a strategy is very positional, an absolute statement for the best strategies is not what we should strive for, all of them are useful if given the right position. Therefore, the teams focused their research on establishing a hierarchy to value each and every move and therefore the value of the position and of the future positions.