Scheduling in job shops with machine breakdowns: an experimental study

This paper considers the simulation-based analysis of dispatching rules for scheduling in dynamic job shops taking into account interruptions on the shop floor. With respect to flowtime and due date-based objectives, the relative performance of well-known, recently proposed as well as some new dispatching rules is evaluated for different settings of the model parameters. The results of the simulation study reveal that the relative performance of scheduling rules can be affected by changing the levels of the breakdown parameters. For the standard model, where all machines are continuously available, as well as for the models taking into account breakdowns of machines, it is shown that for minimizing mean flowtime the performance of one recently proposed rule (PT+WINQ) is significantly better than the performance of all other rules. Analogously it is shown that for minimizing maximum flowtime, and for minimizing variance of flowtime, one new rule (AT−RPT) is superior to all other rules. With respect to due date-based objectives the relative performance of the analyzed scheduling rules is more sensitive to the percentage of time the machines have failures and the mean time to repair.