The abstract should summarize the contents of the paper in short terms, i.e. 150–250 words. Disaster relief operations involve assisting victims in their recovery from either natural or human-caused disasters. In Malaysia, frequent floods pose a recurring challenge, highlighting the importance of efficiently selecting distribution centres for relief efforts, especially given their annual occurrence. A critical aspect of these operations is the strategic choice of operational centres for distributing necessities to victims. However, it is recognized that not all disaster areas can be covered during recovery efforts. To address these challenges, an optimization algorithm is necessary to enhance the coverage of disaster areas. In this context, the K-Nearest Neighbour (KNN) algorithm is employed as a fitness function in both the Genetic Algorithm (GA) and Simulated Annealing (SA). This application aims to optimize the distribution of food from selected distribution centres (DCs). The experiment utilized demand points (DPs) and DCs identified by researchers, adopting a hybrid approach of GA and SA. A performance comparison indicated that the hybrid GA-KNN approach yielded the most optimal solution, with an average fitness value 21% lower than that of the hybrid SA-KNN. This study significantly contributes to the identification of optimal distribution centre locations, ensuring a nearly equal distribution of DPs in each selected location. Such optimization facilitates efficient real-world aid distribution in disaster areas, addressing both time and cost concerns.