Joshua Waring

ITS 360 Hmwk #7

Due: 3/29/19

 

What Would You Do?

3)  I would definitely bring this case to the attention of my firm’s management, as it is the right (and legal) thing to do.  I would explain the error/mistake to them, and also give them a proposed solution to fix the problem.  I would recommend that the software also be immediately updated to remedy this error, and those affected by it be contacted asap.

4) I would recommend to my firm’s management that we delay the launch of the product by another 3 weeks, which should give enough time to fix one of the key features, and be able to provide a patch to fix the second key feature the week after launch.  Unless the software was absolutely dependent on the second key feature to be useable, I think this would be the best way to balance customer satisfaction with the product and sales.  Selling an unfinished product with too many errors could lead to long-term loss of [some of] our customer base, which could be potentially more damaging than a temporary loss in sales caused by the delayed launch of our flagship product.

 

Case Studies

1.1)  It appears that LM’s F35 is now [for the most part] combat ready.  The first fighters have been delivered and deemed-combat ready, and the projected cost of the fighters has dropped now that a finished product is feasible.

This was not always the case.  The Joint Strike Fighter program, which was developed to create and design a fighter to replace many of the US’s current military jets, started in 1996.  Since then it has cost over $300 billion to develop a plane that can be used concurrently by multiple branches of the military.  Issues with this “one-size-fits-all” approach were the main reason it has taken so long and cost so much to develop the F35. 

Lockheed Martin itself has been a major player in the issues plaguing the F35.  For instance, they should have started development on the targeting and radar systems software sooner, and ensured they hired computer programmers with sufficient knowledge to design and implement these systems.  Lockheed-Martin also dealt with substantial lay-offs of their employees, which increased delays in both the design and completion of the F35.  Lastly, the early versions of the F35 that Lockheed-Martin shipped to the military were incomplete and have had to undergo significant upgrades and renovations to bring them up to the standards of the current Joint Strike Fighter program.

1.2)  I think the F35 can be deployed once any serious known bugs are fixed.  However, I believe it should be tested extensively before the military accepts the product from Lockheed-Martin.  Even if there are minor software issues discovered after deployment, I think Lockheed-Martin should be held responsible for fixing those and providing software and security updates for the lifetime of the plane to ensure pilot and operational security.

1.3)  My only real suggestions would be to ensure that proper thorough testing is done before the “finished” product is accepted by the military, and that Lockheed-Martin is held responsible for fixing any software glitches or bugs.  I think that the contract for the military’s purchase of fighter jets should be clear in its time, cost and quality assurance expectations, and the military shouldn’t be afraid of holding its contractors responsible for delays or inadequate products.

2.1) I think that self-driving car accidents are generally the driver’s fault.  Despite the fact that the crash may have been caused by an issue with the autonomous system, current “self-driving” cars are pretty clear with specifying that the driver needs to be ready to intervene at any time.  Because of this, I think the manufacture cannot be held responsible in most cases for crashes of their autonomous vehicles, unless the crash is something caused entirely by the autonomous driving system and could not have been corrected [whatsoever] by the driver of the car.  I think insurance for self-driving cars will cost more at first due to the potential risks involved, but eventually as the systems become reliable, insurance costs will probably go down.

2.2) I think cross-manufacture algorithm-sharing would be beneficial and potentially lead to better (safer) results sooner than if the manufacturer’s kept this information to themselves.  However, logistical sharing and compilation of the merged information would be somewhat difficult and the companies would have to figure out how to “eat” this cost in order to make it worth their while.  I think car manufactures probably would not accept this mandate, as each one hopes to be the first to develop a safe and reliable car, and they know they can make more money if they hold the monopoly on the autonomous vehicle market.

2.3) I think the degree of care should definitely become much greater as the level and complexity of the autonomous software systems increases.  The more autonomous the vehicles, the more sophisticated software is required, and the more potential for bugs or life-threatening glitches.  Hence, I believe quality assurance and degree of care should be upheld to higher standards as level of autonomy in the driverless car industry increases.