Issue link: https://iconnect007.uberflip.com/i/340751
July 2014 • The PCB Magazine 79 WHY REMOvING YOUR BOTTLENECK IS A BAD IDEA continues (above) this blue line. The reason why the criti- cal WIP is four is because this is the point that the maximum theoretical throughput with zero variability is attained. The purple line represents the best case sce- nario for cycle time, which is defined as the total process time plus the total wait time, which is the total time it takes a penny to travel through the entire process. It is impossible to have a cy- cle time that is better (below) the purple line. Is the "best case" bottleneck rate and "best case" cycle time ever achievable in reality? It is not, because of variability. One way to under- stand this inefficiency is to imagine a freeway completely packed with cars touching. Noth- ing is going to move! Having cars touching can only happen if all the cars are going at a 0 velocity with no variation (which is a long- winded way to explain parked cars). This would be the case even if you slipped the transmission from park to drive and slammed down on the accelerator. Lots of burnt rubber, but no move- ment. In order to move we have to have some space between the cars (a buffer); therefore, the number of cars per minute that pass under a bridge is going to be less than the theoretical number of cars that could pass under the bridge if we could all drive at exactly 55 mph. Mark Spearman and Wally Hopp derived a marginal case in order to evaluate the goodness of a specific production system. In Figure 2, the marginal case for throughput is the red line and Figure 2: comparison of the best and marginal cases for throughput and cycle time, with two plots for throughput where the scale is on the left and two plots for cycle time where the scale is on the right. in order to improve significantly over the marginal case (red line for throughput and cyan line for cycle time) you have to do things that go against lean such as unbalancing the line and/or adding capacity.