Project Abstract

Behavioral analysis of data in process systems is the ultimate application of my accumulated experiences in statistics,
mathematics, and physics. The processes themselves are derived from mathematical and physical models, and are streamlined
using optimization methods that are at the heart of calculus. As the theoretically formulated processes are turned into production,
statistical analysis is used to isolate and perfect different aspects of the procedures to ensure the utmost efficiency of the overall
system. Research in process engineering is essential to improving industrial practices. In the case of plastic injection molding,
inefficiencies in production process result in the disruption in the uniformity of the product, excess wastage of materials, and other
potential disasters related to inconsistent measurements and calibrations of machines. The use of data driven methods to analyze
the entirety of the production process assists with the perpetual task of improving and optimizing the outputs while reducing errors
and eliminating opportunities for failure.

Wednesday, March 1, 2017

First week at Microtech Southwest

February 13th was my first day at Microtech Southwest, the plastic injection molding facility where I am doing my project. The first week was all about getting to know all the parts of the fabrication process, as well as familiarizing myself with the various tasks and duties that needed to sustain production. I spent most of my time the first week in the quality lab, where finished parts are sorted, measured, tested, and verified to ensure that each batch of battery seals confirms to all required standards. from each batch of battery seals produced, one seal from each mold is taken out. That set of seals is measured for height, width, hub height, hub depth, and potentially other factors, depending on the model of the seal. One set of seals is also put through a burst tester, to see test the pressure tolerance of the seals. In a battery, the aqueous compound inside the cell can expand with use, and if the seal is breathable enough to maintain equilibrium inside the cell while allowing enough air through so that the cell doesn't explode, then it is perfect. The burst tester applies air pressure to the seals to see if they will fail under the similar conditions, and the data from the burst machine is yet another part of the quality assurance process. The last step of the quality assurance process is the statistical analysis of the data received from all the measurement machines. If the data is within an acceptable margin of error, the seals from that batch are accepted and can then be shipped off to the battery companies where the seals are then used to build batteries.

I also spent some time getting to know the other parts of the factory, mainly one of the vision systems. The vision system is a computer with several cameras connected to it. There is also a belt on which parts are fitted on by a machine, and the parts are passed through the cameras to measure similar values such as height, width, circumference, etc. The computer then processes the data from the cameras, and if the measured values are unacceptable the parts are knocked off the belt into a scrap pile. This machine operates at an extremely fast pace, with the ability to process up to 2000 parts per minute. This means that maintenance of the machine is of utmost importance because one error can cause hundreds of parts to be incorrectly sorted, which could mean that many good parts are rejected, wasting materials or that many bad parts are accepted, destroying the established quality standards. 

Recently, I have been away travelling, and due to the lack of stable internet access I haven't been able to post anything for the past couple weeks. During this temporary break from my time at Microtech, I have been doing some background research on both the plastic injection molding process and on the industrial statistical process, using materials borrowed from Mr. Clark, the director of Engineering at Microtech. I will follow up with a reflection on my findings in another post within the next week.