semi合集-English.pdf - 第3426页

SEMI E133-0705 © SEMI 2004, 2005 7 9 Process Control Sy stems Functional Group Ca pabilities and Interface Data Structures 9.1 Functional Group Partitioning 9.1.1 Process control systems are partitioned into functional g…

SEMI E133-0705 © SEMI 2004, 2005 6

the following criteria are applied to test whether or not a given set of capabilities constitutes a valid PCS functional

group. Specifically, a functional group should satisfy one or more of the following requirements:

 Support a coherent set of tasks

 Share common set of data and must use it under similar constraints (performance, retention, etc.)

 Internal objects are tightly coupled

 Legacy implementation/integration constraints exist (for example, the capabilities are inseparable because an

external system must interact with them as an integrated whole)

 Strong associations exist in the market (that is, one would expect these capabilities to be provided as an

integrated product)

8.2 Based on theses criteria, and the stated implementation priorities of the semiconductor manufacturers and

suppliers involved on the task force, the following initial PCS functional groups have been identified:

 R2R (Run-to-Run) Control

 FD (Fault Detection)

 FC (Fault Classification)

 FP (Fault Prediction)

 SPC (Statistical Process Control)

8.3 Note that it is understood that other PCS functional groups can be added as needed. Future versions of the

document will describe procedures for adding a new group.

8.4 The aforementioned groups can be implemented at the tool or factory level or both. As a requirement,

functional group interfaces should be defined in a consistent way. The following diagram represents the proposed

functional groups and an example of their interactions within the factory and tool environments.

PCS Functions in the “Equipment”

R2R FD FC FP SPC ...

Equipment Specific

Tool Control System

- Process Modules

- Integrated Metrology

- Sensors

- User Interface

- E-Diagnostics

- ...

PCS Functions in the “Factory”

R2R

R2R FD FC FP SPC ...

Factory Specific

Manufacturing Execution Systems

- Recipe Management

- Dispatch

- Automated Material Handling

- E-Diagnostics

- ...

PCS Functions in the “Equipment”

R2R FD FC FP SPC ...

Equipment Specific

Tool Control System

- Process Modules

- Integrated Metrology

- Sensors

- User Interface

- E-Diagnostics

- ...

Figure 1

PCS Functions

8.4.1 Process Control Systems may consist of a variety of capabilities or functions at both the factory and

equipment level as shown in Figure 1. Moreover, this may include multiple instances of the same function type

from different suppliers. In the example above, several run-to-run controllers are running at the factory level with

differences that are only visible as internal special capabilities. This interface standard will facilitate easier

collaboration between the tool level functions and factory level systems independent of the supplier that developed

the internal function’s capabilities.

SEMI E133-0705 © SEMI 2004, 2005 7

9 Process Control Systems Functional Group Capabilities and Interface Data Structures

9.1 Functional Group Partitioning

9.1.1 Process control systems are partitioned into functional groups using the criteria defined in §8. This standard

defines the interface data structures, services and behavior associated with these functional groups. In this section

the capabilities of the PCS functional groups are defined in more detail. First of all, the basic capabilities common

to all

functional groups are presented. Then the additional basic capabilities unique to each functional group are

presented. A high level description of the specifications of data structures, services and behavior are described in

§10.

9.1.2 It is important to note that the specific implementations of these various functional groups may differ greatly

in the level of capability delivered, performance, quality of service, and internal architecture. Moreover, in a given

factory, there may be multiple instances of the same functional group running concurrently, from one or more

suppliers. Some functional groups may be completely independent of other PCS functional groups (in other words,

they can operate in relative isolation, depending only on information from systems external to PCS); others may

depend heavily on other functional groups, collaborating closely to provide a higher level of capability than they

could on their own.

9.1.3 The goal of the PCS interface standards is to enable all of these implementation scenarios (and more) without

dictating internal product architectures.

9.2 PCS Analysis Engine

9.2.1 As defined in §8, all functional groups can be thought of as specializations of a PCS Analysis Engine and

therefore inherit the structure and behavior of a PCS Analysis Engine. This section defines the common capabilities

of a PCS Analysis Engine.

9.2.2 General Capability Description

9.2.2.1 All PCS Analysis Engines shall have inputs and outputs as shown in Figure 2. This means that the PCS

Analysis Engines shall support the consumption of the specified inputs and production of the specified outputs as

necessary to achieve the required capabilities of that analysis engine as specified in this standard. These inputs and

outputs are specified as aggregate data structures. The structures are aggregated into “collection types” according to

the category of data. A definition of these collection types is provided in Table 4. The PCS Analysis Engine shall

have the general behavior of utilizing specified inputs, performing tasks based to some degree on these inputs, and

generating outputs. Any additional structure and behavior of various PCS Analysis Engines are outlined in the

remainder of this section and specified in detail in ¶10.4.

9.2.2.2 Note that, in addition to limitations of specification defined in §4, the following are beyond the scope of

definition of the PCS Analysis Engine:

 the level of synchronization of inputs with performance of tasks and/or generation of outputs, and

 the specification of any internal states of the PCS Analysis Engine.

9.2.2.3 All PCS Analysis Engines belonging to this functional group shall have the capabilities listed in Table 5.

SEMI E133-0705 © SEMI 2004, 2005 8

Figure 2

PCS Analysis Engine Common Input/Output Structure

9.2.2.4 Note that based on the simple data model shown in Figure 2, Inputs are available as an Output from an

Analysis Engine. Data provided as input to an Analysis Engine should be accessible as output from the Analysis

Engine for historical purposes.

Table 4 PCS Analysis Engine Inputs and Outputs Collection Type Definitions

Collection Type Definition Comments

Globals

Information that applies globally across an entire job. A lot id, for example could be

specified once as a global so that

it would not have to be specified

with many other data sets.

AnalysisParameters

Information that identifies numeric data used by the Analysis

Engine to perform various calculations.

Equation coefficients,

feedforward data, etc.

CalculatedOutputs

Data representing the results of analysis from an Analysis

Engine.

A calculation derived from

various inputs to the Analysis

Engine indicating wafer state,

process state or model state

information, recommended

machine settings, etc.

LogMessages

Data describing the activity of an Analysis Engine as it

executes its tasks.

Data describing the activities of

the Analysis Engine. Note that

this does not include log

information on the

communication protocol with the

AE. An example that is

acceptable is “Algorithm #3 with

model parameters ‘A’, ‘B’, and

‘C’”.

ModuleData

Information representing the current or last known physical

state of a process module that will be used by an Analysis

Engine to process material.

Machine, or module E10 state,

RF hours, number of wafers

processed, etc.

ProcessData

Information representing the process itself, such as the name

and type of the process and the current or last known process

state based on measurements from sensors of the processing

environment.

Name of the process (CMP), in-

situ process measurements, trace

data, and summary statistics etc.

SubstrateData

Information about the substrate being associated with the

Analysis Engine. This could include data representing a

single wafer, multiple wafers, or one or more lots, as well as

attributes of the material.

Profile or topology data, and

other characteristics measured

physically or electrically.

PCS AE

Inputs

Globals

AnalysisParameters

CalculatedOutputs

Outputs

Globals

AnalysisParameters

CalculatedOutputs

PCS AE

Inputs

Globals

AnalysisParameters

LogMessages

ModuleData

ProcessData

SubstrateData

Outputs

Globals

AnalysisParameters

LogMessages

ModuleData

ProcessData

SubstrateData