This paper is concerned with the shift scheduling problem involving multiple breaks with different durations and multiple break windows for each shift. We have incorporated ideal break periods and ideal waiting time into the original problem previously presented in the literature. As an extension of the implicit integer programming model with a single goal of minimizing the labor cost, we have proposed an implicit preemptive goal programming model involving three goals, which are given in order of their priority levels as follows: (1) minimize the labor cost; (2) maximize the number of employees that receive their breaks at ideal break periods; (3) make the waiting times of the employees between their consecutive breaks equal to the ideal waiting time, i.e. minimize the deviations from ideal waiting time. The ideal waiting time is incorporated into the model implicitly by matching the periods within the break windows. We aim at improving the break schedules through a more sensitive timing of the breaks, without causing an increase in the labor cost. The computational results obtained on randomly generated test problems indicate that the extended model may yield considerable improvement in the break placement.