I need two things for this discussion hw: 1. One post about this question (about 1page)
2. two replies for their posts (1-2 paragraph for each one)
Question:
The course materials lay out a methodical process for design and design evaluation. So how can engineering systems fail with faulty designs? Discuss how it could be from missing one of the key features noted in the course materials. What if the process seemed to be followed to the letter, could a system still fail?
NishitDalal
Engineering systems with faulty designs can fail horrifically. Not only that, if the design is faulty then it will put people at risk of getting hurt. A faulty design usually happens when something is overlooked, or a requirement is not met. This often happens when not fully addressing what's needed in the system, not following the Life Cycle Engineering of a system can lead up to the system being faulty. Now days, a systems lifecycle is shorter than what it typically was back in the old days because of new technologies being used, this can make something simple more complex just so it can keep up with the recent technological evolution.
An example of a faulty design that failed horrifically is the Challenger that broke apart and led to the death of its crew members on January 28th 1986. The faulty design of this was the failure of the O-Rings design. An engineer Roger Boisjoly found out about this flaw three days after the STS-21-C landed. The job of the O-rings was to help withstand the heat of the launch, there are two O-Rings per joint, that way if the first one fails (this means the heat went through and melted the first ring) then the second one would be there to prevent something from going wrong. Roger and his teams finding was that the O-Rings can be ineffective at rooms temperature. Despite trying to stop the launch, it still happened at January 28th 1986 at a cold day, and this caused the O-Rings to fail. If they could fix this key feature, then the design would not have been faulty. This also shows the failure of the T&E/system validation, this is because despite knowing about the O-Rings, NASA overlooked the problem and put people's lives at risk.
Even if a process is seemed to be followed to the letter, a system can still fail. This is because there could be something not accounted for, or something could be overlooked. For example, a walkway could be made with the wrong design so if someone were to step on it, then it would collapse. This has happened too, the Hyatt Regency walkway collapse in July 17th, 1981. Because of the wrong design it killed over one hundred people.
Pan
This week following up with functional analysis we finally start to design a system based on needs and requirements. In other words, it is now more about HOW to build rather than WHAT to do. For this discussion I would like to start with the third posted question: "What if the process seemed to be followed to the letter, could a system still fail?" My answer is "yes". Even if engineers follow all crucial steps to design analysis and evaluation, it is still possible to eventually have a failed system. The reality is cruel. To have a system works as designed engineers have to ensure everything is 100% fine, while a minor mistake is able to break down the system.
How could an engineering system fail? Typically, it is because of missing key features of design analysis. And also, like mentioned above, it just happens even though the design process is theoretically complete. A recent example of failed engineering system is Samsung Note 7. In fact, this phone brings up a tremendous crisis to Samsung's business while it was expected to be the flagship product of the year. According to Reuters Samsung already lost 17 billion US dollars just because of the global sales losses and recall costs, not including the investment of development of Note7 investment. It is so unusual that a giant enterprise like Samsung, whose business is equivalent to one-fifth South Korean's economy, makes itself trapped in this tremendous disaster. There are few steps that Samsung missed during the period of designing and testing according to my research. For example, steps named "select/build model" and "generate data and run model" in the lecture were missed when Samsung made the batteries. Even though the cause is not determined yet, people mostly believe the issue came from battery. At the time when Samsung was designing the model of Note 7, engineers tried to put a higher-density battery in a thinner phone body thinking of the high demand of long-time power supply as well as slim size on market. However, engineers in SDI, a Samsung company that manufactures most of the Note 7 batteries, never developed a model to evaluate performance of the batteries under different circumstances. As a result, they do not have enough data generated to make sure the battery stays fine working in the phone as part of the system. This could also be classified as a missing of Type 1 testing demonstrated in lecture. Another evidence of faulty Type 1 testing is the failure of the module that transmits battery power. This module is considered as another possible reason that burns the phone. Samsung was so eager to make a perfect model that it wanted Note 7 has every advanced feature built in, such as fast charging. But the power transmission module which was built for the new battery installed in a compact phone body has a risk of disabling with the fast charging technology.
It is also said that Samsung noticed the potential issue of Note 7 when the first batch of production coming out. One hundred thousand phones were tested but no problem detected. I would like to believe Samsung did run the Type 3 testing as any other great enterprise would do. Nevertheless, like what I said at the beginning, sometimes a system just failed even if engineers follow the design steps. It is worth mentioning that even at this moment Samsung has not figured out what is exactly the cause of the accident.