Rel.occupation反应的应该是这个singleproc的前面的entity的到达情况,也就是说,只要这个singleproc前面的工站的entity处于可以被移转到这个singleproc的状态时,这个统计量就在工作了,和这个singleproc出于什么状态没有任何关系(无论他是处于empty、blocked、setup或者unplanned等)。! v( I! N# [1 N" L' A, G; x
所以,setting-up和empty和Rel.occupation之间没有任何必然联系,也就不可能相加等于1.
The relative occupation is the capacity-based ratio of the time the object was occupied, not Paused and not Failed to, N, v. c# t5 J0 k% o
the time the object was not paused and not failed. Take a ParallelProc, with four processing stations, that processes 3 x9 { V3 j5 t; t* u2 ~a single MU during the entire statistics collection period. It has a relative occupation of 25%.. S" W3 m, J# }8 w8 y& L+ C) D
啊!我刚算了一下,REL. OCCUPATION是把加工时间跟BLOCK时间加在一起来算, # p4 h: [. T, n/ ]5 ^$ B5 S/ G8 d' a4 z N' D& c
以我贴图的例子来说:: v% x( i( w0 L5 C# E4 a/ b
0 N2 ]7 v9 C. xWORKING:34.52% & P$ L9 D2 K- zBLOCKED:3.06%! m: R. H a8 M4 @' k; u8 d4 z
两者相加=37.58%+ [6 b& I1 I. u J
30天的小时数=720小时 & ~7 c4 m) g2 m实际加工时间:360小时- B4 Q: S& m) p; L
4 l" W* D' d; }# e( @; b720*0.3758/360=0.75162 w; d7 Q2 @) c, s