semi合集-English.pdf - 第112页
SEMI E10-0304 E © SEMI 1986, 2004 26 R1-6.3 Example 3 — A simplified example with a discrete timeline is shown in Figure R1-12. In th is example there are three m odules (i.e., M1, M2, M3) and three IPFs (i.e. , IPF1, IP…
SEMI E10-0304
E
© SEMI 1986, 2004 25
R1-5.6.2.8 The coat/develop system is likely to have
several more IPFs defined essentially in the same
manner as IPF1 and IPF2, shown above.
R1-5.7 Unscheduled Downtime versus Total Downtime
— In order to make a distinction between unscheduled
downtime and total downtime for an IPF, the IPF
function discussed thus far shall be executed twice:
once to determine an unscheduled downtime state and a
second time to determine a general downtime state.
R1-5.7.1 The first time the function is executed, each
module variable is set to zero if the module is in an
unscheduled downtime state and set to one otherwise.
If the function evaluates to 0, the IPF is in an
unscheduled downtime state. Each contiguous instance
of process-flow unscheduled downtime is a process-
flow failure. If the function evaluates to 1, the IPF may
be in either a general downtime state or an up state.
M
i-UD
=
0, if module i is in an unscheduled down state
1, otherwise
R1-5.7.2 The second time the function is executed,
each module variable is set to zero if the module is in
either a scheduled downtime state OR
an unscheduled
downtime state and set to one otherwise. If the function
evaluates to 0, the IPF is in a general downtime state. If
the function evaluates to 1, the IPF is a neutral state.
M
i-GD
=
0, if module i is in scheduled down state
OR an unscheduled down state
1, otherwise
R1-5.7.3 To summarize, if IPF(M
UD
) = 0, the process-
flow state is unscheduled downtime (UD). Otherwise, if
IPF(M
GD
) = 0 the process-flow state is general
downtime (GD), and if IPF(M
GD
) = 1, the IPF state is
uptime.
NOTE 8: The process-flow state for an IPF that has at least
one module in an unscheduled downtime state may be either
(1) unscheduled downtime, (2) general downtime, or (3)
neutral. Condition (1) occurs when the unscheduled downtime
module is serial within the IPF. Condition (2) occurs when the
unscheduled downtime module is not serial within the IPF,
but a module (or set of modules) that is serial within the IPF
is in a scheduled downtime state. Condition (3) occurs when
none of the downtime modules are serial within the IPF
regardless of being in a scheduled or unscheduled state.
R1-6 Determining Multi-Path Cluster Tool
States
R1-6.1 The temporal mapping rules for determining
multi-path cluster tool “productive/unscheduled
downtime/neutral” states from module “productive/not-
productive” states and IPF “up/downtime/unscheduled”
states is as follows:
R1-6.1.1 If any module is in the “productive” state,
then the cluster tool is in the “productive” state.
R1-6.1.2 Otherwise, if all of the IPFs are in the
“unscheduled downtime” state, then the multi-path
cluster tool is in the “unscheduled downtime” state.
R1-6.1.3 Otherwise, the multi-path cluster tool is in the
“neutral” state.
R1-6.2 Each contiguous instance of a multi-path
cluster tool unscheduled downtime state is a multi-path
cluster tool failure.
NOTE 9: This logic is similar to the relationship between
module states and the equipment states in SEMI E116, where
the SEMI E116 “busy” state is analogous to the multi-path
cluster tool “productive state,” the SEMI E116 “blocked”
state is analogous to the “unscheduled downtime” state, and
the SEMI E116 “idle” state is analogous to the “neutral” state.
However, the actual relationship between these two sets of
states is neither trivial, nor direct.
NOTE 10: Productive time may not be easily allocated to
IPFs in the general case. It is expected that some modules will
belong to more than one IPF and that it will be prohibitive to
ascertain the specific flow for each module from moment to
moment. Hence, metrics that depend on productive time may
not be calculated for IPFs.
NOTE 11: For the metrics presented in this document, it is
not necessary to calculate downtime states other than
unscheduled downtime at the multi-path cluster tool level.
The effect of states other than productive and unscheduled
downtime on the multi-path cluster tool is evaluated as the
aggregate effect of such states on the IPFs.
SEMI E10-0304
E
© SEMI 1986, 2004 26
R1-6.3 Example 3 — A simplified example with a
discrete timeline is shown in Figure R1-12. In this
example there are three modules (i.e., M1, M2, M3) and
three IPFs (i.e., IPF1, IPF2, IPF3). IPF1 uses all three
modules, whereas IPF2 does not use M3, and IPF3 does
not use M2. The state value functions for the IPFs are
IPF1 = M1 × M2 × M3, IPF2 = M1 × M2, and IPF3
= M1 × M3, respectively. Temporal mapping of the
IPFs and the multi-path cluster tool is described in
Table R1-4.
NOTE 12: The condition between time t = 4 and t = 5, where
the multi-path cluster tool is productive but all of its IPFs are
down, is expected only as a transient condition in actual
practice. Once the productive module finishes its task in
progress, the multi-path cluster tool would be expected to go
down. However, while the module is productive, it also is
accruing theoretical processing time used in the SEMI E79
metric, rate efficiency. The allocation of this transient
condition to productive time is necessary in order to guarantee
that rate efficiency evaluated over any interval is never
greater than 100%.
5
100
M1
M2
M3
IPF2
IPF3
IPF1
time
Cluster Tool
Module
S
EMI E10
States
Process
Flow
Up/Down
States
=
P
roductive = Unscheduled Down =
N
ot-Productive
Figure R1-12
Mapping Cluster-Tool States, Example 3
Table R1-4 Temporal Mapping of Example 3
Time Module Events IPF Events Cluster Tool State
t = 0 M1 is productive. All IPFs are up. Productive
t = 2 M2 goes down.
M1 is still productive.
IPF1 and IPF2 go to unscheduled
downtime.
Productive
t = 4 M3 goes down.
M1 is still productive.
All IPFs are in unscheduled downtime. Productive
t = 5 M1 exits productive state.
No modules are productive.
All IPFs are still in unscheduled downtime. Unscheduled Downtime
t = 7 M2 come back up.
M3 transitions into scheduled
downtime.
IPF2 is up. At least one IPF is up.
IPF1 and IPF3 are down, but not in
unscheduled downtime
Neutral
t = 9 M3 comes back up.
All modules are up.
All IPFs are up. Neutral
SEMI E10-0304
E
© SEMI 1986, 2004 27
R1-7 Intended Process-Flow and Multi-Path
Cluster-Tool Metrics
R1-7.1 Once the IPF and multi-path cluster tool state
histories have been prepared, evaluation of metrics may
be performed in a similar manner to that used for
evaluating non-cluster or single-path cluster tools or
modules. The metrics are defined below with sample
calculations based on Example 3 from Section R1-6.
NOTE 13: The multi-path cluster tool metrics presented here
reflect the reliability of the multi-path cluster tool against total
failure rather than partial failure. It is recognized that these
metrics proposed here will reflect more favorably on systems
with higher levels of redundancy at each process step. The
cost or “trade-off” of this redundancy may be evaluated using
other metrics from SEMI E10, SEMI E79, SEMI E35, or from
non-standard evaluations. It also is recognized that these
metrics will not reflect the partial loss of throughput for an
IPF that is “up,” but some of its alternative modules are
“down.” Once again the analyst is encouraged to consult other
metrics from SEMI E10, SEMI E79, or from non-standard
evaluations.
R1-7.2 Multi-Path Cluster Tool Reliability
R1-7.2.1 Multi-Path Cluster Tool Mean Productive
Time Between Failure (MTBF
p-CT
) — mean productive
time between failure where productive time occurs
when at least one module is in the productive state, and
a failure occurs when there are no available IPFs
through the multi-path cluster tool due to module-level
unscheduled downtime.
M
TB
F
p-CT
=
=
=
5 hours
N
umber of Multi–Path Cluster Tool Failures
M
ulti
–
Path Cluster Tool Productive Time
5 hours
1 failure
R1-7.2.2 Multi-Path Cluster Tool Failure Time
(MFT
CT
) — mean time when there are no available IPFs
through the multi-path cluster tool due to module-level
unscheduled downtime.
M
F
T
CT
=
=
=
2 hours
N
umber of Multi–Path Cluster Tool Failures
M
ulti
–
Path Cluster Tool Unscheduled Downtime
2 hours
1 failure
NOTE 14: Since multi-path cluster tool productive time and
multi-path cluster tool unscheduled downtime are mutually
exclusive; they may be compatible with approaches based on
renewal cycle models. Other renewal cycle results (e.g., the
limiting probability of finding the cluster-tool productive or
failed when approaching it at random) similarly may apply, as
determined by the analyst.
R1-7.3 Multi-Path Cluster Tool Availability
R1-7.3.1 Multi-Path Cluster Tool Aggregate IPF
Uptime (Uptime
CT-IPF
) — availability of the multi-path
cluster tool as a function of module
“up/downtime/unscheduled downtime” states and is
evaluated as the aggregate uptimes of all IPFs.
43.3%
10 + 10 + 10
3 + 5 + 5
13
30
×
100
×
100
Σ
all IPFs
IPF Uptime
Uptime
CT-IPF
=
=
=
≈
Σ
all IPFs
IPF Operations Time
×
100
NOTE 15: For reference, the value of the aggregate
availability efficiency metric from SEMI E79 for the same
example is 20/30 or 66.7%. This difference clearly
demonstrates that depending on which combination of
modules is “down,” the effect on the multi-path cluster tool’s
availability may be substantially different than the aggregate
module availability.
NOTICE: SEMI makes no warranties or
representations as to the suitability of the standards set
forth herein for any particular application. The
determination of the suitability of the standard is solely
the responsibility of the user. Users are cautioned to
refer to manufacturer’s instructions, product labels,
product data sheets, and other relevant literature
respecting any materials mentioned herein. These
standards are subject to change without notice.
The user’s attention is called to the possibility that
compliance with this standard may require use of
copyrighted material or of an invention covered by
patent rights. By publication of this standard, SEMI
takes no position respecting the validity of any patent
rights or copyrights asserted in connection with any
item mentioned in this standard. Users of this standard
are expressly advised that determination of any such
patent rights or copyrights, and the risk of infringement
of such rights, are entirely their own responsibility.
Copyright by SEMI® (Semiconductor Equipment and Materials
International), 3081 Zanker Road, San Jose, CA 95134. Reproduction o
f
the contents in whole or in part is forbidden without express written
consent of SEMI.