semi合集-English.pdf - 第588页

SEMI E79-0304 © SEMI 1999, 2004 22 RELATED INFORMATION 2 RAPID CHARACTERIZATION OF IN TRINSIC EQUIPMENT EFFICI ENCY (IEE) AND THE PRODUCTIVI TY EFFICIENCY PLANE NOTICE : This related information is not an official pa rt …

100%1 / 7923
SEMI E79-0304 © SEMI 1999, 2004 21
• Nominal parameters are more representative of the
physical systems being studied.
• Nominal parameters are more representative of the
desired system performance.
• It is not practical to obtain reliable data.
R1-3.3.3 Parametric Models — For cases where the
time for an operational element may have a range of
values that are dependent on recipe specifications,
theoretical time is best represented by a parametric
model. Parametric models for representing
semiconductor operations may be based on
mathematical formulas, e.g., implant time versus beam
current, and/or “lookup” tables, e.g., best observed etch
time vs. etch end point.
R1-3.3.3.1 For the photolithography example, one of
the recipe parameters is the exposure energy (EE).
Given the ideal or theoretical lamp intensity (LI) of the
stepper, the theoretical duration per exposure (THT
EX
)
for the recipe may be calculated as THT
EX
= EE / LI .
SEMI E79-0304 © SEMI 1999, 2004 22
RELATED INFORMATION 2
RAPID CHARACTERIZATION OF INTRINSIC EQUIPMENT EFFICIENCY
(IEE) AND THE PRODUCTIVITY EFFICIENCY PLANE
NOTICE: This related information is not an official part of SEMI E79. This related information was approved for
publication by vote of the responsible committee on December 15, 1999.
R2-1 Rapid Characterization of Intrinsic
Equipment Efficiency
R2-1.1 IEE may be rapidly characterized as follows
using a limited number of production experiments.
R2-1.2 Step 1: Design Production Experiments —
Select a limited number of scenarios to execute as
equipment experiments. It may be of interest to
characterize IEE according to either operating modes,
processing diversity, or a combination of operating
modes and process diversity.
R2-1.2.1 Determine the value-added in-process
theoretical production time per unit (VTHT) – (time per
unit) for all recipes involved. This information is
required for determining IEE. See Table A2-1.
R2-1.2.2 For assessment of operating modes,
experiments should examine only a single typical recipe
that is likely to be used most frequently. Each
experiment should examine a separate operating mode,
where examples of various equipment operational
modes may be as follows:
One wafer at a time mode (production monitor
wafer).
One batch is run, then the tool stops.
Two batches are run, then the tool stops.
Three batches are run, then the tool stops.
Continuous (cascade) mode. A larger number of
batches is run, then the tool stops.
R2-1.2.3 For assessment of process diversity, select a
limited number of representative recipes that will be
processed by a tool. This population should include at
least one recipe representing the minimum expected
processing duration and one representing the maximum.
R2-1.2.4 Each experiment should be designed and
executed to eliminate rate efficiency losses to the
greatest extent possible. It is further assumed that
quality efficiency losses are zero. Under this
approximation, (VA-OEE) = (OEE)
× (IEE).
R2-1.3 Step 2: Execute Process Experiments —
Perform process experiments recording all relevant
input variables including the configuration of lots and
wafers, and recipes. For each experiment, record the
elapsed production time using convenient means, e.g.,
stopwatch or existing data acquisition system.
R2-1.4 Step 3: Calculate Results — Calculate IEE and
throughput for each experiment. IEE may be used to
measure the effect of non-value-added overhead time
during equipment processing. Approximately:
Intrinsic Equipment Efficiency (IEE)
= (Value-Added In-Process Theoretical Time)
/(Non-Value-Added Overhead Time
+ Value-Added In-Process Theoretical Time)
R2-1.4.1 Hence, non-value-added overhead time for
each experiment may be calculated as:
Non-Value-Added Overhead Time
= [(Production Time)
- ( Value-Added In-Process Theoretical Time)]
R2-1.4.2 It should be the focus of efforts by the
equipment supplier and the end-user to reduce or
eliminate non-value-added overhead time through
improved equipment design, including scheduling
software, as well as hardware components (carrier and
wafer handling systems, valves, pumps, heaters,
coolers, etc.).
R2-1.4.3 Results may be shown in either tabular form
or plotted graphically on a Productivity Effectiveness
Plane. See Section R2-2.
SEMI E79-0304 © SEMI 1999, 2004 23
Sample Rapid Characterization of Intrinsic Equipment
Efficiency (IEE)
Four experiments were designed and executed with the
following results:
Exp.
Units of Recipe i
Per Experiment
Value-Added
In-Process
Theoretical
Production Time
Per Unit
(VTHT
i
)
Production
Time Per
Experiment
1 50 of recipe A 0.00670 hr/unit 3.3333 hr
2 100 of recipe B 0.00550 hr/unit 5.0000 hr
3 150 of recipe B 0.00550 hr/unit 6.0000 hr
4 300 of recipe A 0.00670 hr/unit 10.0000 hr
Effective Unit Throughput Per Experiment
= (Total Units Per Experiment)
/(Production Time Per Experiment)
Intrinsic Equipment Efficiency (IEE) Per Experiment
= [Σ
i
(Units of Recipe i Per Experiment × VTHT
i
)
/(Production Time Per Experiment)] × 100%
Non-Value-Added Overhead Time Per Experiment
= [(Production Time Per Experiment)
- Σ
i
(Units of Recipe i Per Experiment × VTHT
i
)]
Experiment
Effective
Unit
Throughput
Intrinsic
Equipment
Efficiency
Non-Value-
Added
Time
in Hours
1 15 units/hr 10.05% 2.9983 hr
2 20 units/hr 11.00% 4.4500 hr
3 25 units/hr 13.75% 5.1750 hr
4 30 units/hr 20.10% 7.9900 hr
R2-2 Productivity Efficiency Plane (PEP)
R2-2.1 It is recognized that OEE and throughput are
separate metrics whose relationship may not be
straightforward. Because theoretical production time
per unit may vary widely by recipe, good throughput
performance may not indicate correspondingly good
performance in terms of overall equipment efficiency.
Similarly, a high OEE score may not be indicative of
high throughput. Given this disparity, it is essential that
both metrics be analyzed and compared as separate
entities.
R2-2.2 Data from rapid characterization experiments
may be displayed graphically on a Productivity
Effectiveness Plane (PEP) diagram, Figure R2-1.
R2-2.3 In PEP diagrams, IEE is plotted as a function of
throughput. Points that appear further to the right have
higher throughput and points that appear higher up have
higher IEE. It is desired to have combined equipment
and process designs whose performance would appear
in the upper right quadrant of the plane for the entire
operating range of the tool.
R2-2.4 The four experimental data points from the
sample problem are plotted in Figure R2-1 and, when
connected, appear to approximate an upward sloping
curve. This curve is referred to as a tool signature.
R2-2.5 Tool signatures may be used to describe tool
performance relative to isolated variables. This
representation helps equipment suppliers and users
visualize the effects of tool operating modes on IEE and
throughput. For more complex multi-dimensional
experiment sets, the tool signature would appear as a
hyper-surface.
R2-2.6 OEE can be plotted for comparison against tool
signatures. In Figure R2-1, an OEE measurement of a
typical week is plotted. For the known throughput
corresponding to this OEE score, the value of IEE on
the tool signature for the same throughput may be used
to approximate the IEE score for the week without
calculating IEE explicitly.