Design improvement of screw down stop valve assembly through K Means clustering algorithm

Design improvement is becoming vital for enhancing product performance, reducing costs and meeting customer needs and market demands. Recent advancements utilize clustering algorithms for supporting informed decision making by grouping similar objects based on common characteristics. Among other clustering techniques, K-Means clustering is widely used because of its simplicity and scalability which makes it ideal for a wide range of applications. The screw-down stop valve controls the fluid flow by adjusting the position of the valve seat attached to the stem within the valve body. Minimizing manufacturing costs and assembly time across diverse sectors is crucial to improve competitiveness. This study is focused on improving the design of the valve assembly without affecting its functionality. The K-Means algorithm grouped components based on their material similarities and interactions, and these were combined using an interaction based merging method. The structural integrity of the re-designed components was validated through a static structural analysis. The modified product was fabricated through additive manufacturing using polylactic acid material to verify its manufacturability and form. The findings indicated that the maximum equivalent stress obtained for both the modified design segments were found to be 51.82 MPa and 53.7 MPa which is well within the yield strengths of their respective material. Additionally, the number of components in the modified product was reduced from 14 to 5.