The selection of a “best team” is arguably one of the more important aspects of preparing for any event. In cricket, a sport that is watched by nearly 2.5 billion persons, this is quite a complex task. Ensuring the most optimal skill sets are complemented by the most capable players is integral to a team successfully completing a match as the victors.

Selecting the West Indies Cricket team for a particular tournament is a challenging task further complicated by the relatively large number of aspirants, occasionally exceeding 500 aspiring team members. The current process of selecting the playing 11 for a match is currently predominantly dependent on a Selection Board, a group of individuals tasked with analysing each aspirant's historical performance to determine whether the player should be added to the team being composed.

This is a heavily data-driven process which can be exponentially improved both in accuracy and efficiency with the employment of Artificial Intelligence (AI) and Machine Learning algorithms. With this in full purview, a methodology is being developed using statistical data on various historical matches acquired from cricinfo.com. The data is analysed and a model developed in accordance with the Knowledge Discovery and Data Mining (KDDM) process model. This model is then used to provide guidance to Machine Learning algorithms, resulting in the most optimized teams being selected with particular enforced constraints.

Advances of the use of aspects of AI in sports, though relatively young, has made significant feats in most recent years. Automated team selection for a cricket team is a substantially underdeveloped area of study to date. This research seeks to build a Decision Support System which will aid the decision makers to select a squad of players with the highest probability of dominating a One Day International (ODI) tournament.