The present era demands the efficient modelling of any manufacturing system to enable it to cope with unforeseen situations on the shop floor. One of the complex issues affecting the performance of manufacturing systems is the scheduling of part types. In this paper, the authors have attempted to overcome the impact of uncertainties such as machine breakdowns, deadlocks, etc., by inserting slack that can absorb these disruptions without affecting the other scheduled activities. The impact of the flexibilities in this scenario is also investigated. The objective functions have been formulated in such a manner that a better trade-off between the uncertainties and flexibilities can be established. Consideration of automated guided vehicles (AGVs) in this scenario helps in the loading or unloading of part types in a better manner. In the recent past, a comprehensive literature survey revealed the supremacy of random search algorithms in evaluating the performance of these types of dynamic manufacturing system. The authors have used a metaheuristic known as the quick convergence simulated annealing (QCSA) algorithm, and employed it to resolve the dynamic manufacturing scenario. The metaheuristic encompasses a Cauchy distribution function as a probability function that helps in escaping the local minima in a better manner. Various machine breakdown scenarios are generated. A ‘heuristic gap’ is measured, and it indicates the effectiveness of the performance of the proposed methodology with the varying problem complexities. Statistical validation is also carried out, which helps in authenticating the effectiveness of the proposed approach. The efficacy of the proposed approach is also compared with deterministic priority rules.