2005年01月19日

Master's Thesis Complete

I have finished my master's thesis and submitted it to the department. I am so glad to put this behind me and look ahead to the future. This has been haunting me for eight months now, since I "finished" the first edition last April. Since then I had done more research, and had to update several chapters. It was funny, the due date was last Friday, and my advisor sends me a message Thursday night saying, "Your thesis needs work." and then submits comments to me on Saturday. I guess a due date doesn't mean all that much here at MIT. It just defines a range in which you can turn it in. ;)

The topic you ask? The title is "Cost Estimation of Functional and Physical Changes Made to Complex Systems." It basically comes up with a method for estimating the development cost of design changes at different phases during the design. I threw in some other stuff on key cost drivers and traceabilty. I've learned a lot about cost estimation, system engineering, and the aerospace industry by finishing this thesis. Though they are doing great things (see last post), for the most part design is done haphazardly (design, build, test, break, fix, build, test, break, fix, ...). This is why these aerospace projects tend to cost tax payers so much, is because everyone in the industry optimizes the HECK out of everything. This causes unecessary iteration in the development phase, and an amazingly complex system that is expensive to operate during its life-cycle. Changes must be made to the industry, but it can only be done with the new wave of engineers graduating from universities. Design is not learned from experience, it must be done in a logical way that minimizes cost. So quit introducing changes, and Keep It Simple Stupid (KISS).

If you would like to read the thesis, you can find it here. The abstract and introductory chapter are good, and the rest may be a little too steep in details for most.

January 19, 2005 05:12 PM