Lean Six Sigma Process Improvement at Headwall Photonics

Hess, Kaitlyn; Brooks, Carson O.; Bradley, Johann S.; Boermeester, Caitlin N.; Legatt, Kayla A.

Student Work

Lean Six Sigma Process Improvement at Headwall Photonics

Public

The purpose of this MQP was to evaluate one of Headwall Photonics current cleaning processes and present recommendations for process improvements at their manufacturing facility in Fitchburg, Massachusetts. The objective of this project was to reduce the scrap rate, improve repeatability, and reproducibility in the pre-clean process for Headwall Photonics. The rationale behind this was that Headwall’s current pre-clean process was inconsistent and experiencing high scrap rates. The methods utilized began with collaboration from subject matter experts at Headwall to better understand and define company parameters of what to optimize in the cleaning process. Scrap rate, reproducibility, and repeatability were three key parameters identified to improve in the cleaning process. Additionally, when examining the whole process of manufacturing gradients, an improved design for fixture was identified as a way to mitigate scrap rates. Due to the fragile nature of these products during manufacturing, a standardized process that is poka-yoke (“mistake-proof”) in its design is critical in order to mitigate waste in the process. The project’s first step to understanding how to achieve our objective was to use axiomatic design (AD) decomposition. Further, generating a value stream map of the process to identify bottlenecks and assemble a value-effort graph allowed us to choose favorable alternatives to improve the given optimization parameters. Next, our team directly simulated Headwall’s current pre-clean process in a lab-based environment at WPI in order to compare their current process with a proposed alternative that may improve scrap rates. The success of each cleaning process was evaluated through contact angle measurements and cost-benefit analyses. One of the methods explored included the use of carbon dioxide, while another utilized plasma technology to clean the surface of the substrate. The results of our experiments with varying the pre-clean process were that the CO2 process was more efficient but the results were not consistently better than Headwall’s original process using an acetic bath. The plasma cleaning experiment is still underway at the Plasma Institute at Drexel University, but initial evaluation indicates that this method could potentially prove beneficial to Headwall’s pre-clean process if they chose to incorporate it. Additionally, the new fixture design for the final efficiency test incorporated a spring mass system that will decrease the chance of defects at the step in the process. This fixture was developed through a series of design iterations and prototyping. It resulted in an improvement from the previous design in that it reduces the chance of an operator damaging the part during loading and unloading of the substrate. To conclude, our team recommended Headwall continue with their current acetic acid process but incorporate lean process improvements to enhance effectiveness and efficiency. This project also identified a new cleaning method, plasma technology, that could potentially be used by Headwall Photonics in the future. Lastly, a series of recommendations was made, including an improved process map and work instructions, that will improve performance for the pre-clean aspect of Headwall’s manufacturing process.

This report represents the work of one or more WPI undergraduate students submitted to the faculty as evidence of completion of a degree requirement. WPI routinely publishes these reports on its website without editorial or peer review.

Creator

Publisher