This document summarizes a project to reduce shell weight for BH521 6-cavity shells from 27kg to 25kg while maintaining low defect rates. Key actions included measuring current weights, analyzing factors like investment time and temperature, conducting experiments to optimize parameters, implementing controls at 65-70s and 240°C, and verifying the new process achieved an average weight of 25.5kg. Savings of 1.4kg resin per shell and an extra shell per hour were projected to yield annual savings of Rs. 5.88 lakhs. Controls were established for materials, times, and temperatures to sustain the improvement.

1. Shell Weight reduction in
BH521 6 Cavity
Duration Sept‐2007 to Nov‐2007
Project Leader – Abdullah Ansari

2. 2
Project Start date : 01/09/07
Target Completion date : 15/11/07
Date of proj presentation : 20/11/07
Throughput Project Open days : 75
Sponsors reviewed proj : Yes
No of team meetings : 06
Savings : Current FY Rs. 2 L
Annualized Rs. 6 L
Phase completed : D M A I C
Belts % time on project : 40%
Project review status : ( R / Y / G )
Sponsors MBB
Define
Measure
Analysis
Improve
Control
GGGGG
GGGGG

3. Define project objective, expected benefits to the organization, team
members, project schedule.
DEFINE
3

4. 4
Objective:
Benefits:
Project
Team:
To increase number of BH521(6 Cavity) shells per MT of resin coated sand by
reducing shell weight from 27 Kgs to 25 Kgs at molding while maintaining the
minimum possible defectives due to shell making.
S J Chougule (Prod Engg)
Savings: 52.5 K Rs per Month
Improved process cycle time
Reduced operator fatigue
Effective utilization of raw material
A T Ansari
Ashok Khanna S S Kulkarni (Prod Engg)
R K Choudhary Mahesh (Proc Engg)
S N Soddy (Quality)
R K Jena (Melting)
Schedule: Measurement: 20/09/07 Analyze: 10/10/07
Improvement: 10/11/07 Control: 15/11/07
Revision 1
Shell Weight reduction in BH521 6 Cavity

5. Base Current Shell Weight – 26 to 27 Kgs Current
Current Cycle Time – 257 Seconds per shell (14 shells per hour)
Investment time – 85 Seconds
MEASURE
5

6. 6
50,000 ft Process Map
Shell Weight reduction in BH521 6 Cavity
Inputs (Xs) Outputs (Ys)
C/U/S
Resin Coated sand C No. of Shell per hour
Cycle Time C Shell weight
Heated pattern U
C = Controllable
U = Uncontrollable
S = SOP
Shell Making

7. 7
C & E Matrix
Rating of
Importance to
Customer
8 8
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Process Step Process Inputs
No. of Shell
per hour
Shell weight
Total
1 Shell Making Investment time 6 6 96
2 Shell Making Resin coated sand 0 6 48
3 Shell Making Preheat time 2 2 32
4 Shell Making Preheat temp 0 2 16
5 Shell Making Baking time 2 0 16

8. 8
‘I’ Chart for Shell Weight before improvement
I Chart of Shell Weight
1 7 13 19 25 31 37 43 49 55
28.5
28.0
27.5
27.0
26.5
26.0
25.5
Observation
Individual Value
UCL=28.247
_
X=26.882
LCL=25.517
1

9. Which factor affecting on shell weight?
ANALYZE
9

10. 10
Histogram of Shell Wt.
25.8 26.4 27.0 27.6 28.2
20
15
10
5
0
Shell Wt
Frequency
Mean 26.88
StDev 0.4979
N 60
Histogram of Shell Wt
Normal

11. 11
Regression Analysis
Shell Weight S 0.494345
155 160 165 170 175 180
28.5
28.0
27.5
27.0
26.5
26.0
Pattern Temp
R-Sq 3.1%
R-Sq(adj) 1.4%
Fitted Line Plot
Shell Weight = 24.35 + 0.01501 Pattern Temp
Pattern temperature is not the major factor resulting in increase or decrease of shell weight.
Take trials by varying temperature and investment time and redo regression analysis.

12. 12
Regression analysis – Fitted line plot for Inv time
65 70 75 80 85
27.5
27.0
26.5
26.0
25.5
25.0
Investment time
Shell weight
S 0.195085
R-Sq 96.5%
R-Sq(adj) 95.6%
Fitted Line Plot
Shell weight = 18.58 + 0.1025 Investment time

13. To decide optimum process parameters to achieve required
shell weight.
IMPROVE
13

14. 14
DOE
RUN Pattern Temp x Investment T ime Plan / Status Actual Avg Temp
Actual Avg Shell
Weight
Run 1 220 x 65 04.10.2007 220.8 25.14
Run 2 220 x 75 04.10.2007 220.9 26.21
Run 3 220 x 85 04.10.2007 220.9 27.08
Run 4 240 x 65 07.10.2007 240.8 25.04
Run 5 240 x 75 07.10.2007 240.2 26.13
Run 6 240 x 85 07.10.2007 238.8 26.63
Run 7 260 x 65 08.10.2007 260.1 25.36
Run 8 260 x 75 08.10.2007 262.6 27.09
Run 9 260 x 85 08.10.2007 261.4 28.04

15. Contour Plot of Shell weight vs Pattern Temprature, Investment time
Shell weight
< 25.0
–
25.0 25.5
25.5 –
26.0
26.0 –
26.5
26.5 –
27.0
27.0 –
27.5
27.5 –
28.0
15
A
B
AB
DOE Results
0 5 10 15 20
Term
Standardized Effect
1.99
Factor Name
A Inv estment time
B Pattern Temprature
Pareto Chart of the Standardized Effects
(response is Shell weight, Alpha = .05)
Investment time
Pattern Temprature
260
250
240
230
220
210
65 70 75 80 85
>
28.0
Investment time = 69.9450
Pattern Temprature = 220.231
Shell weight = 25.5168
Investment time = 65.0255
Pattern Temprature = 240.176
Shell weight = 25.2558

16. 16
Optimized Process controls for further full heat trials
• Investment time = 65 to 70 Seconds
– Target 67/68 Seconds
• Pattern Temperature = 230 to 250 0C
– Target 240 0C

17. 17
Results of full heat trials with new process parameters
1 5 9 13 17 21 25 29 33
26.4
26.2
26.0
25.8
25.6
25.4
25.2
25.0
Observation
Individual Value
UCL=26.362
_
X=25.651
LCL=24.940
I Chart of Shell Wt
23.6 24.0 24.4 24.8 25.2 25.6 26.0
30
25
20
15
10
5
0
Shell Wt
Frequency
Mean 25.00
StDev 0.5079
N 172
Histogram of Shell Wt
Normal
Data of 1st
Trial
Data of shell weight 172
shells ~ 5 heats

18. 18
Any effect on Casting Rejection…?
• Before
• Casting rejection 6%
• Casting rejection due
to swelling and shell
crack – 1.5%
• After
• Casting rejection in
full heat trials 5%
• Casting rejection due
to swelling and shell
crack – 1%
No Negative Effect

19. 19
Improvement Summary
• Earlier shell weight 26.9Kgs (based on data collected on 18/09/2007), Now
25.5Kgs
• Saving of 1.4 Kg resin coated sand per shell
• Earlier investment time 85/90 Seconds, Now 67 Seconds
• Cycle time reduction by 18/23 Seconds per shell.
• Forecasted saving
– Due to saving in resin coated sand
• Rs 49000 per month
• Rs 5.88 Lacs per annum
– Productivity improvement
• 20 seconds x 14(shells per hour) = 280 Seconds ~ 1 shell per hour
• Capacity enhancement of 3744nos of castings per month

20. Introduce Controls to sustain the improvement
CONTROL
20

21. 21
Control Plan
• Resin coated sand
– Sand properties should be strictly within specifications at receiving
inspection.
– Build up should be 49% to 52%
• Investment time
– Should be 65 to 70 Seconds. 67 is ideal. To be ensured while setup
approval.
• Pattern Temperature
– Target for 2400C – Measure by infrared gun while setup approval
– Maintain temperature by keeping burners on/off for every 2 alternate
shells, as temperature on machine can not be controlled
automatically.

22. Thank You