The inherent dynamic nature of industrial environments often needs not only the cyclic revision of scheduling decisions (typical rescheduling actions) but also an efficient adjustment of the production recipe to the current process conditions. Therefore, the concept of flexible recipe becomes an important part of the rescheduling framework that allows full exploitation of the process flexibility in batch plants. This work introduces a rigorous mathematical approach that incorporates the concept of recipe flexibility to plantwide batch operation rescheduling. The proposed mixed-integer linear programming (MILP)-based approach is able to address the rescheduling problem of multistage, multipurpose batch plants involving different storage policies, nonzero transfer times, and flexible batch product recipes. This model relies on the concept of general precedence, which reduces the number of binary variables and, therefore, the computational effort. Flexible-recipe constraints are incorporated in this model to account for the possibility of changing the processing time of some tasks, tweaking the rest of the parameters of the product recipe. The cost for modifying these process variables from their optimal economic conditions is taken into account to represent how productivity is increased despite the cost of altering the nominal plant conditions. Different incidences, such as insertion of new orders, equipment failures, due-date changes, maintenance tasks, and delay in arrivals, variations in the cost, and quality of the raw materials or products, taking place throughout the scheduling horizon, are considered to evaluate the effectiveness of the proposed approach.