Implementing statistical process control in Yarn Manufacturing (Spinning)

Abstract

Introduction: One of the major problems of Indian spinning mills is inconsistency in the yarn quality. Cotton, the raw material, being a natural fibre has inherent variations. The process adds its own variability. However, the market demands high and consistent quality. Hence there is a need to minimise process variations through process control and continuous process improvements. The foundation of implementing statistical process control SPC has been laid by undertaking this project at Excel Spinning Limited. Objectives of the project: Within limited time available and as desired by the management following objectives wereset.1. To minimize the count variation of final product - yarn.2. To check and analyse the consistency of carding quality in terms of neps.3. To implement control chart practices. Methodology: Past and present data for the requisite attributes were collected from the mill records and by collecting the samples where necessary. Appropriate control charts were developed and process capability was established. Brainstorming, cause and effect analysis and controlled trials were carried out to identify the source of variations. Findings: The source of count variation could be traced. The back process machine contribution to count variation could be traced. It was established by the trials that controlling the output quality of draw frame and comber process could minimize count variation. Analysis of card nep data showed a cyclic pattern throughout the period of one year. This is in relation with the qualities of fibre blends used in the cotton mixing and major maintenance programs carried out during the period. Implementing control chart practices were initiated but could not be continued due to lack of constraints like unavailability of the staff, computing facility and involvement of the middle management staff. Conclusions: The outcomes and the report were discussed with the mill management. For important fibre properties, in process and final yarn quality attributes that are regularly tested to be plotted on control/run charts. To start with, even if statistical theories are not applied the results can be plotted on a graph paper where the sudden fluctuation can be identified easily and action can be taken only when it is necessary. This would avoid unnecessary corrective actions applied only by seeing numerical values, which disturbs the process rather than correction in many cases. The employee involvement needs to be started which is lacking. To begin with small cross-functional groups can be formed. After identifying their training needs, appropriate training has to be given to understand and apply SPC techniques for minimizing process variations and continuous process improvements. This can be extended to all sections of operations once confidence is developed and results are seen. Operating practices like blending of sliver cans, piecing of sliver, block creeling to mention few needs to be uniformly followed and is possible through involvement and training of low level employees. Quality based rewards/incentives for outstanding performance, suggestions that leads to quality improvement and cost savings to be initiated. Even though the findings and recommendations appear simple, effective implementation to the recommendations could lead substantial gain in long run. Quality improvements in term would lead to cost reduction. SPC could be a stepping stone for re-vitalizing the organisation towards TQM.