The automotive industry faces a rising demand for customized vehicles, which requires the use of Mixed-Model Assembly Lines (MMALs). However, these lines struggle with uncertainty in process time, making production planning challenging. To solve this issue, the proposed solution is to use a Markov Decision Process (MDP) to model the uncertainty in process time, task assignment, and workers' movements dynamically. We formulated this line balancing problem as a Mixed-Integer Linear Programming model and solved it using a Benders decomposition algorithm. The approach demonstrates significant improvements over the traditional approach currently used in the automotive industry.