Wright Williams & Kelly
Factory Explorer® v2.3
Bias in Simulation-Based WIP Estimates Summary: For models with many ramp points, Factory Explorer's simulation-based WIP estimates presented on the Tool Group Results Worksheet may be biased low. There is no simple workaround for this problem. This error is fixed in Factory Explorer v2.4. Details: When running a simulation, at each ramp point, Factory Explorer rebuilds queue length and WIP information for all tool groups. This rebuild is necessary, since the tool used for a process step may have changed. The rebuild is accomplished by zeroing out queue length and WIP statistics for all tool groups, removing queued lots from all queues, re-adding lots to the appropriate tool group queues, and then recalculating queue length and WIP statistics. This mechanism, however, does not properly account for lots being processed at the time of the ramp point, leading to WIP statistics that are biased low. Due to the nature of this error, bias is most likely to occur in models with many ramp points. Four columns on the Tool Group Results Worksheet are affected: "Sim Avg Unit WIP", "Sim Max Unit WIP", "Sim Avg Lot WIP", and "Sim Max Lot WIP". In Factory Explorer v2.4, the code has been modified to correctly account for lots in process at the time of the ramp point. Workaround: There is no simple workaround for this problem. The impact of the problem is limited, however, to the four columns on the Tool Group Results Worksheet identified above. Techalert tafx17-9806
© 1995-2003 Wright Williams & Kelly, Inc.
|
