Reduction of Paint line Defects in Shock Absorber Through Six Sigma DMAIC Phases

Abstract

In modern era, the Six Sigma tools and techniques have been implemented in various manufacturing sectors, which strive to ameliorate continuous improvement in achieving less variation, cost and high quality of end products. This case study narrowly focuses on reduction/elimination of two imperative responses in spray painting process producing shock absorbers, namely peel off and blisters using the Six Sigma Define-Measure-Analyze-Improve-Control (DMAIC) approach that highly impacts quality at customer end. The define phase rolls out the tools such as Pareto chart, voice of business (VOB) and project charter which identifies pretreatment in the spray painting process as the critical stage. The measure phase reveals the continuous assessment of spray painting process, with intense brainstorming sessions the imperative responses were culminated as peel-off and blisters. In analyzing phase, the vital root causes that impact the responses were identified as cleaning temperature, phosphate temperature and phosphate pH (power of Hydrogen) by using cause and effect diagram and Likert scaling. The improve phase concentrates on optimizing the vital root causes which impact the responses by using Taguchi robust design approach. The L27 orthogonal array (OA) had been constructed with three factors and levels,results of experimentation had been analyzed by using Analysis of Variance (ANOVA) and multivariate regression which identifies the condition of optimality on peel off and blisters in the pretreatment process. In control phase, the confirmation run with optimality conditions were conducted, the results obtained from runs are satisfied which embarks the sigma level from 3.31 to 4.5. The continuous pursue on eliminating variation in the processing stage was attained by framing a control plan to control the variation within acceptable levels in the pretreatment process.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 54 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.