Applications of Monte-Carlo Tree Search in Software Engineering

Industry 4.0 encourages the use of Artificial Intelligence in industry problems such as the development and quality of Software. Monte-Carlo Tree Search (MCTS) is a search algorithm based on a random sampling of the search space that has been used in games like Tic Tac Toe, Chess, Rubik’s Cube or Sudoku. The objective of this Systematic Literature Review (SLR) is to identify opportunities to generate lines of research of this search algorithm, identifying the applications of the MCTS, as well as its advantages and disadvantages compared to other search algorithms in the Software Engineering. In the results obtained in this work, some applications of MCTS in software engineering were found, specifically in software testing and Software Product Line (SPL). It was concluded that MCTS, despite being a relatively new algorithm, can be applied to some areas of Software Engineering. As future work, it is proposed to investigate its application in other areas and compare it with the current methods of Search-Based Software Engineering.