Conflicts occur naturally in the real world at all levels of society, individually, in groups or society as a whole. Almost all the existing conflict resolution models are unilateral in their decision-making process. They do not consider the actions of the involved parties simultaneously. Therefore, in this paper, we aim to design a novel conflict resolution model based on game-theoretic rough sets by constructing a game between all the concerned parties (players), computing the payoff of different strategies and classifying them following equilibrium rules. The proposed model yields more realistic and accurate results as it explores all possibilities and is flexible in determining different threshold values relative to the complexities of real-life problems. Three real-life conflict situations are solved with the proposed model, and a comprehensive analysis is done to validate the effectiveness of the proposed approach.