A SIX SIGMA APPROACH FOR OXIDATION DEFECT REDUCTION IN GTAW OF A NUCLEAR PLANT

Transcription

1 A SIX SIGMA APPROACH FOR OXIDATION DEFECT REDUCTION IN GTAW OF A NUCLEAR PLANT 1 SHANMUGA RAMAN G., 2 VENKAT RAMAN B., 3 KUPPUSAMY M V 1,3 Research Scholar, Sathyabama University, Chennai, India 2 Associate Director-RSEG, Indira Gandhi Centre for Atomic Research, Kalpakkam, India Scientific Officer / E, Indira Gandhi Centre for Atomic Research, Kalpakkam, India 1 ggshanmugaraam@yahoo.co.in Abstract --Nuclear safety has acquired a renewed interest in the wake of Fukushima catastrophe. Every process of a nuclear plant, whether operational or project based, however minor it be, needs to be perfected to ensure safety and reliability. This paper details out the approach adopted in improving the reliability of a nuclear piping facility in India, by reducing the Oxidation defects of GTAW process. The proven DMAIC methodology is used for identifying the improvement strategies. The findings of the research suggests adoption of two innovative alternatives to internal back purging and also highlights importance of welders skill in ensuring elimination of Oxidation defects in GTAW. Absence of literature on application of Six Sigma in nuclear industry and the challenge of adapting the DMAIC approach with the right tools provided the motivation for the research. Index Terms-- Six Sigma, DMAIC, DPMO, GTAW, Weld Defects, Oxidation, Nuclear piping facility I. INTRODUCTION A. Six Sigma Sigma is a statistical concept that represents the amount of variation present in a process relative to customer requirements or specifications. The term Six Sigma refers to a business philosophy of focusing on continuous improvement by understanding customers needs, instituting proper measurement methods and analyzing business processes. Since its inception at Motorola in the beginning of 8s Six Sigma has been applied successfully in numerous manufacturing and non-manufacturing organizations having mass repetitive processes. However, very little research has been carried out in applying Six Sigma in a comprehensive manner in welding process. Six Sigma advocates two methodologies: DMAIC (Define, Measure, Analyse, Improve and control) and DMADV (Define, Measure, Analyse, develop and Verify). DMAIC methodology is used for improving the current capabilities of existing processes and DMADV approach is used for creating new process/product/service to meet customer requirements. This study adopts DMAIC methodology for applying Six Sigma in the GTAW process of a nuclear research organization in its nuclear piping facility fabrication for reducing Oxidation defects. Five phases of DMAIC methodology: Define the problem, the process, the CTQ factor (Y) and Xs influencing the CTQ and the specific project goals. Measure current defect level and collect data on the possible Xs. Analyze the data to investigate and verify cause-and-effect relationships between the Xs and the Y. Improve the process by identifying and implementing solutions for the defects. Control the improved process to prevent it from falling back to previous status. B. NUCLEAR RESEARCH ORGANISATION This research was carried out at one of the demonstration nuclear plant of a nuclear research organization in India, which involves 62 kilometers of stainless steel (ASTM A312 TP 34L) web of nuclear piping welded by GTAW (Gas Tungsten Arc Welding) designed to carry highly corrosive nuclear fluid. This paper is the third in the series of studies carried out focusing on Oxidation defects as the Lack of Fusion defects were covered in an earlier study. B.1 GTAW Gas tungsten arc welding (GTAW), also known as tungsten inert gas (TIG) welding, is a welding process which uses a constant current power, tungsten electrode to produce arc for the weld and a filler metal. Fig.1 gives the process setting of GTAW. (4) Fig.1 GTAW process setting GTAW produces exceptionally good quality welds and is used to weld thin sections of stainless steel and Proceedings of 25 th IRF International Conference, 1 th May 215, Chennai, India, ISBN:

2 A Six Sigma Approach For Oxidation Defect Reduction In GTAW Of A Nuclear Plant non-ferrous metals. The process grants the operator greater control over the weld than competing processes such as shielded metal arc welding (SMAW) and gas metal arc welding (GMAW), allowing for stronger, higher quality welds. However, GTAW is very complex slower process calling for greater dexterity and skill on the part of the welder. GTAW finds application in producing high quality joints in the nuclear, aircraft and food industries. In the nuclear piping facility under study, the dense web of pipes spanning over an altitude of 13 meters and the resulting congestion and abnormal positioning of the welder make the GTAW welding even more complex and demand ambidexterity on the part of the welder. B.2 Oxidation Oxidation is a defect that occurs in GTAW around the weld when it is exposed to oxygen in the air.(fig.2) Oxidation is intolerable for nuclear application which requires good corrosion tolerance in demanding service conditions coupled with radioactive atmosphere. Fig. 2 Oxidised Terminal Joint Fig.3 Process flow chart of piping fabrication CTQ (Critical To Quality) characteristics of the process was understood through discussions with the client / sub-contractor and the Quality assurance Division. Absence of discontinuity and non-thinning of material in the piping is critical as the corrosive nuclear fluid travels through it. Absence of noncontinuity in the piping structure becomes all the more important as no repair work can be attempted after commissioning of the plant. As the possibility of occurrence of non- continuity is higher in weld joints, defect free GTAW welds were identified as the CTQ of the piping plant. B. MEASURE Data on the defects encountered during the fabrication of the piping facility was collected from 1316 sample weld joints and the base line sigma level was established. DPMO (Defects Per Million Opportunities) was calculated using the formula (1). II. APPLICATION OF SIX SIGMA A. DEFINE The process of fabricating the ss piping facility was studied and a process flow chart was made (Fig.3) As there were 9 possible GTAW defect categories (Table 1) and as there were 439 defects that have been encountered in the sample welds chosen for the study, the DPMO was calculated to be 4728 and the corresponding Sigma Level was read from the table and established as 4.1 sigma which offers ample scope for improvement. Table 1: GTAW Weld defect Categories Proceedings of 25 th IRF International Conference, 1 th May 215, Chennai, India, ISBN: In order to identify vital few defects from the trivial many, defect categories were prioritized using a Pareto chart based on the sample data (Fig.4).

3 Count A Six Sigma Approach For Oxidation Defect Reduction In GTAW Of A Nuclear Plant Pareto Chart of Weld Defects Defects LF OXI TI U/C LP POR EP CON DIS Count Percent Cum % Fig.4.Pareto chart of GTAW defects As it may well be seen from the Pareto chart, it is the OXI (Oxidation) that has the highest occurrence accounting for 21% of the total defects, leaving out LF which has already been dealt with by a separate study and has been covered in a separate research paper. An exhaustive list causes for Oxidation were identified through a brain storming session and the same are depicted in the form of a Cause and Effect diagram (Fig.5 ) Percent a) Congestion & inaccessibility b) Ineffective purging c) Type of Joint (Terminal Vs Flow) d) Skill of the welder There are totally 35 weld joints in the demonstration plant. It was decided to collect data from the Weld Inspection reports and the radiography reports of weld joints on all the variables from a sample of 1316 weld joints using systematic sampling method. Data on welders experience, their qualification and their migration status were collected from welders database. C. ANALYSE Collected data on the above variables have been analyzed with appropriate statistical tests using Minitab. The tables and output of statistical tests are given for each of the variable. C1. Migration factor Table 2 : Oxidation by Migration Cause & Effect Diagram for OXI Environment Material Personnel Draughty Conditions OXI 2-Sample % Defective Test for OXI - Migrated vs Direct Summary Report Measurement Methods Insufficient purging duration Insufficient gas shield while cooling. Machines Fig.5 Cause and effect Diagram of Oxi B.1 Root Cause Analysis Root cause analysis was done to spot the significant few causes from among the exhaustive list of causes depicted in the Cause & Effect diagram. The causes were evaluated by their chance of incidence in the fabrication environment at IGCAR, given the effective and exacting Quality Assurance Plans backed by well defined and tested procedures for every activity and stringent stage inspections during fabrication. Probability of occurrence of the different causes were estimated through an opinion survey among the experts of GTAW at IGCAR on a rating scale of 1-1, where 1 was most likely root cause, 5 represented may have an influence but not likely to be root cause and 1 was least likely. Based on the probability rating, it was decided to consider the four causes having a score of 7 and above as significant root causes. Yes.1 P =. The % defective of Migrated is significantly greater than the % defective of Direct (p <.5). -2 %Defective Test Is Migrated greater than Direct? 9% CI for the Difference Does the interval include zero? >.5 No 2 Statistics Migrated Direct Total number tested Number of defectives % Defective %CI (1.57, 2.55) (.43,.72) Difference in %defectives * % CI (.97, 1.94) Comments -- Test: As the 'P' value is. which is less than.5 it can be concluded that the %defective of Migrated is greater than Direct at the.5 level of significance. -- CI: Quantifies the uncertainty associated with estimating the difference from sample data. You can be 9% confident that the true difference is between.97 and Fig.6 Two Sample % defective test of OXI by migration As the P value is. which is less than.5 It can be concluded that the % Oxidation of Migrated is greater than Direct at the.5 level of significance and there is association between migration status of welders and the quantum of Oxidation defects caused by them. Proceedings of 25 th IRF International Conference, 1 th May 215, Chennai, India, ISBN:

4 C2. Experience of welders in GTAW A Six Sigma Approach For Oxidation Defect Reduction In GTAW Of A Nuclear Plant Table 3: OXI defects by experience of welders Fig.8 Output of Two Proportion test by Qualification Table 5: OXI defects by type of joints Chi-Square % Defective Test for OXIDATION by Welders' Experience Summary Report Yes.5.1 Do the %defectives differ? P =. Differences among the %defectives are significant (p <.5). >.5 %Defectives Comparison Chart Red intervals that do not overlap differ. No Which %defectives differ? # Welders' Exp Differs from 1 4 to 6 year > 8 to 1 ye 4 3 > 6 to 8 yea < 4 years > 1 years 4 to 6 year Comments > 8 to 1 ye > 6 to 8 yea < 4 years > 1 years As the 'P' value is. whicj is <.5, it can be concluded that there are differences among the %defectives at the.5 level of significance. Use the Comparison Chart to identify % defectives that differ. Red intervals that do not overlap indicate %defectives that differ from each other. Consider the size of the differences to determine if they have practical implications. Fig.9 Output of Two Proportion test by joint type Fig.7 Chi-Square % defective test by experience As the P value is. it can be concluded that there is an association between Experience of welders and quantum of OXI defects. C3. Educational Qualification of welders Welders with 1 th qualification and ITI qualification are performing the GTAW welding at the Nuclear piping facility. Analysis of the OXI defects caused by welders with 1 th and ITI qualifications revealed a significant association between qualification and quantum of OXI defects as seen from the table No.4 and the result of the two proportion test. (Fig.8) Table 4 : Oxi defects by Qualification of welders As the P value. is less than.5, it can be concluded that the occurrence of Oxi defects in Terminal joints are greater than in Flow joints. D. IMPROVE Chi-square test and Two-Proportion tests reveal that Skill of the welder reckoned in terms of experience, qualification and migration does have a statistically significant relationship with the quantum Oxidation defects caused. High degree of skill in maneuvering the welding torch is required as there is congestion and inaccessibility of weld joints and the consequent need for welder to position himself abnormally while welding. Joint type does have strong influence on the quantum of Oxidation defects as inert gas purging for GTAW of pipes become ineffective especially in the cases of Terminal joints, Hookup joints end closure joints and wherever joints are inaccessible due to complexity of system design. Hence the researcher has suggested that welders with ITI qualification and with minimum of 4 years of experience in GTAW should be engaged as they have the skill of maneuvering the welding gun to ensure proper shielding gas flow to the pipe joints. Proceedings of 25 th IRF International Conference, 1 th May 215, Chennai, India, ISBN:

5 A Six Sigma Approach For Oxidation Defect Reduction In GTAW Of A Nuclear Plant It has also been suggested to develop and adopt Perforated Flask Technique to purge terminal joints for ensuring proper purging.(fig.1) Usage of Perforated Flask Technique will also reduce the quantum of Argon Gas usage by half and result in considerable saving. Fig.1 Perforated Flask arrangement For weld joints which are inaccessible, usage of gas forming flux covered (FC) GTAW welding filler materials is suggested. This will eliminate internal purging of Argon gas for ss pipe root welding and will also reduce incidence of Oxidation because of improper purging E. CONTROL Implementation of the above suggestions in the functional plant will result in the reduction of Oxidation defects level. Sustaining the improvement is of equal importance as otherwise the process may slip back to the earlier sigma level. The revised welder eligibility criteria in terms of Experience, qualification and Migration factor should be incorporated in the contract specifications and the same has to be monitored when new welders are inducted in to the team or when new contractors are introduced. Oxidation defect percentage has to be monitored on a monthly basis using C chart to ensure and sustain the reduced Oxidation defect level. CONCLUSION The outcome of the study validates the relevance of Six Sigma and its DMAIC methodology in a noncontinuous and non-repetitive activity like GTAW process. When the suggested improvement strategies are implemented during the fabrication of the functional plant, it is expected to bring about considerable reduction in Oxidation defects and the consequent improvement in safety, reliability and cost savings in terms of rework and avoidance of project overruns. ACKNOWLEDGEMENT The corresponding Author would like to express his gratitude to Dr.B.Venkatraman, Associate Director- RSEG, IGCAR, Kalpakkam, Dr.A.Ravishankar, Associate Director- RPSD, IGCAR, Kalpakkam and Mr.M.V.Kuppusamy Scientific Officer, QAD, IGCAR Kalpakkam for their invaluable support and guidance in executing this project. REFERENCES [1] Snee, R.D., Six Sigma: the evolution of 1 years of business improvement methodology, International Journal of Six Sigma and Competitive Advantage, 24, Vol.-1 No. 1, pp [2] Pyzdek, T. (1999), The Complete Guide to SixSigma, Quality Publishing, Tucson, Az. [3] Harry, M. J; Schroder, R; (2), Six Sigma; The Breakthrough Management Strategy revolutionizingthe world s top Corporations, Doubleday, New York [4] ebsco.org/gtaw Proceedings of 25 th IRF International Conference, 1 th May 215, Chennai, India, ISBN: