Task elimination may actually increase throughput time

The well-known Task Elimination redesign principle suggests to remove unnecessary tasks from a process to improve on time and cost. Although there seems to be a general consensus that removing work can only improve the throughput time of the process, this paper shows that this is not necessarily the case by providing an example that uses plain M/M/c activities. This paper also shows that the Task Automation and Parallelism redesign principles may also lead to longer throughput times. Finally, apart from these negative results, the paper also show under which assumption these redesign principles indeed can only improve the throughput time.