Material Flow Simulation of TF Production Lines

± Results & Benefits (Example based on CIGS Turnkey)

Dr. Roland Sturm, acp-IT

Dr. Hartmut Gross, centrotherm Photovoltaics
Jörg Talaga, centrotherm Photovoltaics
Contents

1 Simulation of TF Manufacturing Lines

2 Enabling Factors for a Successful Simulation Project

3 Example Model

4 Simulation Output & Results

5 Benefits from Manufacturing Simulation

3KRWRQµV 4th Production Equipment Conference , 05.03.2009 , Munich 2

Material Flow Simulation of TF Manufacturing Lines

ƒ TF manufacturing lines
high degree of automation for transport & material handling
ƒ Layout: flow line, job shop or mix of both

Flow line with tact Job shop layout with central transport

ƒ Typically a tact time drives a flow line.

ƒ Substrates MOVE, STOP, WAIT, ACCUMULATE
Tact is often interrupted. Æ There is no flow of material!
ƒ Simulation clarifies the timing dynamics in the manufacturing line.

3KRWRQµV 4th Production Equipment Conference , 05.03.2009 , Munich 3

Enabling Factors for a Successful Simulation Project

Fast Model Execution

and Flexible Output
Generation
Short Build-up Time of ‡ Fast Simulation
First Model Execution

Systematic Input Data ‡ Library-based EQ ‡ Time Measurement

Acquisition Components between any Points
in the Manufacturing
‡ Fab Layout ‡ Standardized Process
Interfaces of the EQ
‡ Equipment Key Data Components ‡ Powerful Online-
(Timing Behavior, Statistics
Availability) ‡ Flexibility to Add
Features to EQ
‡ Constraints of Components
Processes
3KRWRQµV 4th Production Equipment Conference , 05.03.2009 , Munich
4
Building-up the Model ± Think in Components!

ƒ Model conception: Separation of the line in components.

ƒ Each line component to be modeled with specific timing behavior
including unscheduled & scheduled down time.
ƒ Consideration of total system availability in a flow line.

Conveyor EQ
& FiLo Buffer
Process EQ Inspection
EQ
Conveyor Conveyor
EQ EQ

3KRWRQµV 4th Production Equipment Conference , 05.03.2009 , Munich 5

Single Point Failures in a Flow Line

Processing  EQ Material  Handling  

&  Transport  EQ
Availability  in  %   96% 99%
Unscheduled  Down  Time  
per  year  in  days 14,6 3,65
80 20
Repair  Time  Distr.

1  h
2  h 40 15
4  h 20 9
8  h 10
24  h 5
MTTR  in  hours 3,61 1,95
MTBF  in  hours 86,7 193,5
Number  of  EQ  in  the  Line 33 140

Average  Number  of  EQ  in  

Down  State  at  One  Specific  Time 1,38 1,41

3KRWRQµV 4th Production Equipment Conference , 05.03.2009 , Munich 6

Maintenance Scheduling & Buffer Management

Buffer 1 Buffer 2 Buffer 3

Process Process Process Process
Segment 1 Segment 2 Segment 3 Segment 4

Process Seg 4

WIP Buffer 3
Process Seg 3
WIP Buffer 2
Process Seg 2
WIP Buffer 1

Process Seg 1

Production Scheduled Maintenance Time

3KRWRQµV 4th Production Equipment Conference , 05.03.2009 , Munich

Buffer WIP ± Theory vs. Simulation

Simulation
Output
Theory
Process Seg 4

WIP Buffer 3
Process Seg 3
WIP Buffer 2
Process Seg 2
WIP Buffer 1
Process Seg 1

Production Scheduled Maintenance Time

3KRWRQµV 4th Production Equipment Conference , 05.03.2009 , Munich

InFrame Synapse Simulation Library
- Rapid Simulation Model Building

InFrame Synapse Simulation Library Contents:

ƒ Processing & Material Handling EQ for PV Manufacturing
ƒ Routing and Dispatching Functions
ƒ EQ State Models (processing, loading/unloading, maintenance, breakdown)
ƒ Reports for Line Performance Monitoring & Online Statistics
ƒ Optional: Interfaces for Communication with Control Systems (MES)

3KRWRQµV 4th Production Equipment Conference , 05.03.2009 , Munich 9

Systematic Input Data Acquisition

ƒ Example: Sputter Process EQ

‡ Capacity: Number of Substrates in the Equipment
‡ Substrate with ‡ MTBF, MTTR
SubstrateID & Size ‡ Maintenance Cycle
‡ Yield Rate

‡ Speed of Sputter Process Pumping Chambers:

‡ Constraint: Closed Carpet in ‡ Capacity
the Process Chamber ‡ Pump Time Æ Tact

3KRWRQµV 4th Production Equipment Conference , 05.03.2009 , Munich 10

Timeline Reports & Cycle Time Histogram

3KRWRQµV 4th Production Equipment Conference , 05.03.2009 , Munich 11

Simulation Results & Input Data for TCO

TCO: Total Cost

ƒ Variability in Material Supply & of Ownership
Safety Levels FC: Fixed Cost
RC. Running Cost
YC: Yield Cost
ƒ WIP Timeline, Inventory Statistics
L: Lifetime
T: Throughput
ƒ Buffer Capacity Y: Yield
U: Utilization
ƒ Required Equipment Capacity
FC RC YC
TCO
ƒ Product Cycle Time Statistics L T Y U

ƒ Line Throughput as Timeline

ƒ Overall Line Yield

ƒ Equipment & Line Utilization

3KRWRQµV 4th Production Equipment Conference , 05.03.2009 , Munich 12

Line Simulation ± Benefits for TF Manufacturers

ƒ Systematic input data collection for engineering team

ƒ Proof of line control policies Æ MES & line controller
ƒ Ramp-up, maintenance scheduling, buffer management
ƒ PUSH and PULL analysis
ƒ Transparency for factory dynamics
ƒ Effect of single point failures
ƒ Verification of buffer capacities Æ investment!
ƒ Proof of capacity profile
ƒ ³'H-%RWWOHQHFNLQJ´
ƒ Required redundancy to guarantee output targets
ƒ Feeding TCO analysis with dynamic data
ƒ Utilization and throughput considering dynamic yield and line dynamics

3KRWRQµV 4th Production Equipment Conference , 05.03.2009 , Munich 13

 Benefits of Manufacturing Simulation

New Technology /
Planning Phase Ramp up Phase Volume Production
Capacity Expansion

Material Supply Line Simulation Detection of Output Estimation

Planning with MES Efficiency in Transition Phase
Production Volume

Safety Levels Shortening Potentials Optimized

MES Integration Cost Reduction Capacity
Early Bottleneck Time Adaptation
Identification,
Line Configuration Appropriate
Throughput Max. Maintenance
Scheduling
Time-to-Volume Verification of
Risk Analysis, Higher Line
Data for Total Cost Efficiency
of Ownership

Time
3KRWRQµV 4th Production Equipment Conference , 05.03.2009 , Munich 14
Thank you for your attention!

We are looking forward to your visit at the exhibition:

PHOTON's 4th Photovoltaic Technology

Show Europe 2009.

acp-IT
Hall C1, M15

centrotherm Photovoltaics
Hall C1, D4

acp-IT AG
advanced clean production Information Technology AG
Handwerkstr. 29, D-70565 Stuttgart

Dr. Roland Sturm

e-Mail: roland.sturm@acp-it.com
Phone: +49 (0)711-7824089-29