Capacity and cycle time-throughput understanding system (CAC-TUS) an analysis tool to determine the components of capacity and cycle time in a semiconductor manufacturing line

A literature review on variability in semiconductor manufacturing: The next forward leap to Industry 4.0

“…As utilization, as well as variability, also creates higher cycle times, one way to visualize the intrinsic variability of a manufacturing system is through "operational curves" or "cycle time throughput curves". These curves are found in 20 of the papers that we reviewed, and used in 12 of them to explain the impact of variability (Brown et al 2010, Delp et al 2006, Etman et al 2011, Ignizio 2011, Kim et al 2014, Martin 1999, Robinson et al 2003, Schoemig 1999, Shanthikumar et al 2007, Tirkel 2013, Wu 2005, and Zisgen et al 2008. These curves represent the Xfactor (the ratio between cycle time and raw processing time) as a function of the utilization of available capacity at different variability levels (as illustrated in Figure 1.A).…”

Cycle Time Approximations for the G/G/m Queue Subject to Server Failures and Cycle Time Offsets with Applications

“…The additional time may be considered as production time and incorporated into the service distribution. By doing so, the capacity loss associated with the idle with WIP, described in [6], is accounted for (increased production time implies increased loading). Combining idle with WIP and the approximation of (1) yields the following approximation for the expected cycle time in a G/G/m queue subject to server failures and idle with WIP: (2) where ρ * = λ(Ω + 1/µ)/(mΑ), 0 < ρ * < 1.…”

“…Standard statistical analysis can be applied to the tool and lot logistics databases in a manufacturing facility to obtain the statistics. IBM's 200mm semiconductor wafer fabrication facility has developed many automated data analysis tools [6,8] to enable the acquisition of needed data. In particular, idle with WIP time, travel time, hold time, post production unloading time, utilization (to some extent) and tool availability are generated automatically for each toolset.…”

“…Section II develops an approximation formula for the mean cycle time in a G/G/m queue subject to server failures drawn from a consideration of the approximations of [2,4]. Here, idle instances in which work is present (idle with WIP, see [6]) are incorporated. The presence of hold time and travel time prior to arrival at the server is considered in Section III.…”

“…The intuitive value of simple expressions for system performance coupled with their ease of use has resulted in their ascendance as the primary tool for understanding cycle time behavior in IBM's 200mm semiconductor wafer fabricator. Much work has been devoted to successfully bridging the gap between measures of actual system performance and the predictions obtained from the M/D/1 (or the more generic G/G/m) queue [6,7,8,9,10,11]. …”