Upload
ngotruc
View
219
Download
1
Embed Size (px)
Citation preview
www.metec-estad2015.com
A Model for Sequencing and
Optimizing Steel Melt Shop
Operations
Using Iterative Hierarchical Decomposition
based Discrete Event Simulation
Steelmaking
The Production Process Flow at the Client’s
Plant
2
Atanu Mukherjee, Arnab Adak
M.N. Dastur & Company
16.06.2015
Blast
Furnace
Basic
Oxygen
Furnace
Basic
Oxygen
Furnace
PIG Casting
Machine
Slag
Dumping
Yard
Ladle
Refining
Furnace
Vacuum
Degassing
Unit
Continuous
Caster
Ho
t m
eta
l la
dle
by lo
com
otive
s
Liq
uid
Ste
el L
ad
le b
y
Tra
nsfe
r C
ar
and
Cra
ne
Liq
uid
Ste
el Ladle
by
Tra
nsfe
r C
ar
and
Cra
ne
Liq
uid
Ste
el L
ad
le b
y
Cra
ne
Sla
g p
ot b
y
Tra
nsfe
r C
ar
Scrap
Yard
Scra
p B
ox
by C
rane
Liquid Steel Ladle by
Transfer Car
and Crane
Ladle
Preparation
Aisle
Empty Ladle
by Crane
Ironmaking
Objective: Overall Throughput Improvement
3
Atanu Mukherjee, Arnab Adak
M.N. Dastur & Company
16.06.2015
Minimize tap-to-tap time of
furnace
Maximize utilization while
maintaining heat sequence
Optimal resource utilization
• Cranes
• Ladle cars
Basic Oxygen Furnace
(BOF)
Ladle Refining Furnace (LRF)
&
Vacuum Degassing Unit (VD)
Caster
Unit Optimization
Operations Synchronization
Overall Throughput Improvement
Bottleneck Analysis Using Operational Laws
4
Atanu Mukherjee, Arnab Adak
M.N. Dastur & Company
16.06.2015
Melt Shop
Operations
Use a queuing
model
Maintain job flow
balance
Apply utilization
law
BOFService Time
LRFService Time
VDUService Time
CasterService Time
Pro
port
ional
The unit with the
highest total
service time can
be a limiting
factor in
achieving higher
throughput
Reduce the cycle time to the minimum level possible at the
bottleneck unit
BOFUtilization
LRFUtilization
VDUUtilization
CasterUtilization
Iterative Cycle Time Optimization
5
Atanu Mukherjee, Arnab Adak
M.N. Dastur & Company
16.06.2015
Reduce the cycle time to
the minimum level possible
at the bottleneck unit
Sim
ula
te th
e m
odel
Again
with
the re
duce
d
cycle
times Stack rank the new
service times of the units
in the production chain
Bottle
neck s
hifts
during
itera
tion
s o
f
impro
vem
ent
Continue iterating
till there is no
further
opportunities for
practical cycle time
improvement
Iterative Capacity Utilization Improvement
6
Atanu Mukherjee, Arnab Adak
M.N. Dastur & Company
16.06.2015
• Further scope for improvement might exist across
production units, service units and buffer units
• Observe the behaviour of capacity utilization of the units
which could potentially create a bottleneck
• Capacity utilization of units might change and tend to
move in a way affecting its performance
• Adding unit capacity and/or buffer capacity can further
improve system performance
Addition of unit capacity needed to be guided by • Economics
• Cost Benefit
Cycle Time
Approach Using Iterative and Hierarchical
Decomposition
7
Atanu Mukherjee, Arnab Adak
M.N. Dastur & Company
16.06.2015
Unit 3
Unit 1
Unit 2
Capacity Utilization
Unit 1
Unit 2
Unit 3
Progressive hierarchical decomposition of the chain and iteration based on
the two heuristics drives the simulation towards an optimal solution
Baseline Simulation
8
Atanu Mukherjee, Arnab Adak
M.N. Dastur & Company
16.06.2015
• One year operation data was collected for
the BOF, LRF, VD and Caster
• Fitted into distributions and best fit
distribution was selected with the help of
goodness of fit tests
• The model was run with the existing
operation parameters
Results
Average heats per day 21
Average BOF tap-to-tap time 65 mins
LRF utilization 79 %
Caster utilization (using 2
strands) 53 %
VD utilization 48 %
Caster heat sequence Maintained
Snapshot of 10 days with existing
parameters after attaining stabilization
BOF Basic Cycle Time Distribution
9
Atanu Mukherjee, Arnab Adak
M.N. Dastur & Company
16.06.2015
Activities Time*
Scrap charging 1 to 2
Hot metal pouring 4
Blowing 16 to 18
Deslagging 2 to 3
Temperature & Analysis 5
Tapping 4 to 6
Slagging off 2
Slag coating 5
Ba
sic
activitie
s in
BO
F
*Figures in minutes
Weibull
Mean:43 ; Standard Deviation: 0.7
Average delta of 22 mins between Tap-to-Tap and BOF basic
cycle time
BOF Cycle Time Improvement Strategy
10
Atanu Mukherjee, Arnab Adak
M.N. Dastur & Company
16.06.2015
Event Average
Time*
Mouth jam cleaning and dozing after
every 8 heats 90
Tap hole changing after about 15
heats 45
Other maintenance after 75 heats 180
Current Maintenance Downtime
Practices
Event Average
Time*
Mouth jam cleaning and dozing after
every 15 heats 45
Tap hole cleaning and other
maintenance after 75 heats 180
Proposed Maintenance Downtime
Practices
*Figures in minutes
Use T
ap h
ole
sle
eves
Improved BOF Tap-To-Tap Time Distribution
11
Atanu Mukherjee, Arnab Adak
M.N. Dastur & Company
16.06.2015
Lognormal
Mean: 59; Standard Deviation: 4.6
Skewness: 1.21585
Although average tap-to-tap time improved to 59 mins,
non trivial positive skew indicates potential BOF
blockage due to downstream
*Figures in minutes
Capacity Utilization Analysis
12
Atanu Mukherjee, Arnab Adak
M.N. Dastur & Company
16.06.2015
Results
Average heats per day 23
Average BOF tap-to-tap time 59 mins
LRF utilization 92 %
Caster utilization (using 2
strands) 60 %
VD utilization 54 %
Caster heat sequence Maintained
Snapshot of 10 days with proposed
parameters after attaining stabilization
• Probable downstream blockage in the
LRF
• Removing downstream blockage can
further reduce BOF tap-to-tap time
• Make use of the blocked BOF capacity
as well as the additional caster capacity
with three strands in operation
?
45 60 65 (time in minutes)
13
Atanu Mukherjee, Arnab Adak
M.N. Dastur & Company
16.06.2015
Lognormal
Mean: 51; Standard Deviation: 0.8
Skewness: 5.93932e-002
Resulting improvement in BOF Tap-To-Tap
Time Distribution
Reduced skewness and more symmetrical
pattern indicates lesser system wide congestion
Additional Capacity along with Cycle Time
Improvements Increases Throughput Significantly
14
Atanu Mukherjee, Arnab Adak
M.N. Dastur & Company
16.06.2015
• Average BOF tap-to-tap could be brought down further to an
average of 51 minutes
• The caster could be operated with three strands while
maintaining the heat sequence
• The overall throughput increased by about 20%
The increased overall steel production decreased the diversion of hot
metal to the pig casting machine improving the profitability of the melt
shop
Cost Benefit Analysis
15
Atanu Mukherjee, Arnab Adak
M.N. Dastur & Company
16.06.2015
Unit Costs & Operational Parameters
Tap hole sleeves 200 mm, 10 sleeve pack $3000
Sleeve change interval 75 heats
Electrode consumption in LRF 12 gm/KWH
Electricity consumption in LRF 0.5 KwH/Degree Celcius/ton
Average heating in LRF 5 Degree Celcius
Cost of Electricity 10 C/ KWH
Investments
1X35 Ton LRF 8 MM$
Product Price
Average Billet Price $480 /ton
Average Pig Iron Price $ 400/ton
Marginal Revenue R 5.3 MM$/year
Marginal Cost C (Sleeve Cost + Electrode
Consumption + Power Consumption) ~$400,000
Additional profit , P 4.9 MM$
Payback Period for Investment 1 Year and 7 Months
Future Work
16
Atanu Mukherjee, Arnab Adak
M.N. Dastur & Company
16.06.2015
Large scale plant wide throughput improvement through
Cycle time reduction
Capacity addition
Include operational elements like intra-plant logistics, facility layout re-engineering,
plant wide inventory & movement buffers, routing sequences and material
allocations
Use hybrid models of discrete event and system dynamic simulation for overall
performance improvement
17
Atanu Mukherjee, Arnab Adak
M.N. Dastur & Company
16.06.2015
M.N. Dastur & Company (P) Ltd Consulting Engineers
www.dastur.com