What is the "Problem of many hands"? Have you come across this problem personally before?
Two definitions of problem of many hands
- It's not easy to pinpoint who to blame
- Multiple people and multiple decisions that lead to the final disaster. Need them all to happen for the whole thing to go wrong
How could a code of ethics be more enforced to stop the "problem of many hands" from happening?
- Acknowledge that mistakes will happen
- Test others work, not yourself
- Always having a redundancy
- Separate ethics committee (reduces conflicts of interest when issues show up)
What are the similarities and differences between the 737 MAX case and other real world engineering disasters (Ford Pinto, Space Shuttle Challenger, Arianne V)?
- Ford Pinto - Exploding gastank
- Similarities - Reputation hit, Management prioritising profit, commercial
- Differences - Cost benefit analysis, risk/frequency, you can still buy one
- Space Shuttle Challenger - Hardware failure
- Similarities - The problem was found but management overruled
- Differences - 1 time thing vs 2 time thing, non commercial
- Arianne V - Reused software which wasn't compatible
- Similarities - Sensor failures
- Differences - reused hardware vs reused software, non commercial
How do the issues in the 737 MAX case differ from the fictional 'Killer Robot' scenario from Week 2? In what ways are they the same?
Similarities - Lack of training, So many errors in a short time, No easy override, commercial
Differences - Involvement of regulatory authorities,
Was sufficient training provided for pilots? Is it ever okay to not disclose potential issues about the system to the pilot in-charge? How does this relate to the concept of informed consent?
There was not sufficient training provided for the pilots, only online iPad training was provided which might have been sufficient if the plane was similar to the previous generation. However in this case, the new MCAS system should have been shown to the pilots and training should have been required for it.
Boeing made these decisions as they would have had to re-certify the plane if they identified it as a new model. In terms of informed consent, Boeing did not make clear that the new 737 MAX was a different aircraft to previous aircrafts which resulted in unqualified pilots.
What ethical concerns are there in the relationship between Boeing and the FAA? What could have been done differently?
- The relationship between Boeing and FAA was "cozy"
- Consequentialism: Boeing and Airbus had a history of creating and certifying working and safe planes. FAA did not believe that the consequences of self-certification were unethical as Boeing had a good track record of ethical behaviour - hindsight bias.
- Duty-based: FAA didn't perform their duty of checking the planes and just relied on the self-certification from Boeing.
- FAA should have required oversight of the self-certification process. FAA should have upheld their ethical guidelines to provide safe aircraft by verifying Boeing's results.
- Boeing should have encouraged an atmosphere of cooperation by allowing engineers to safely report behaviour that they believed was against the code of conduct such as an improper review process.
Did competition between Boeing and Airbus lead to the circumvention of ethical norms ? Is competition inherently unethical?
- Boeing definitely decided to make the new planes after airbus announced their planes
- Competition itself is healthy, but in this case, boeing decided to cut corners, and compete for market share on the cost of safety
- Not training pilots
- Software was used to compensate for hardware inadequacies
- Software was obscure from pilots
What alternate ethical steps could Boeing have taken in order to prevent the disasters ? Justify how these steps are ethical.
- Spending more time
- Proper training
- Proper testing
- Not relying on past successes, as you have a duty to be thorough with testing
- Informing the pilots about the software
- Companies should indulge in practices that ensure action is taken when engineers bring up an issue - engineers here knew of the issue but did not bring it up until after the crash
- At the heart of ethics code of conduct in most industries, is the safety and well being of the users of the product. Inherently engineers are bounded to solve problems whether or not they caused it, and so there were many times where engineers could have stopped this product from going forward and prevent the crashes.
"Problem of Many Hands"
- Hard to identify the root cause of an issue
- Lots of small events that seem insignificant, that culminate and cause a bigger issue
- "Just culture" - Not afraid to bring up issues and problems
The FAA let Boeing self certify themselves.
Many common similarities with other engineering projects
- Ford Pinto - Prioritisation of profit over rather than reputation, safety, etc...
- Space Shuttle "Challenger" - A problem was identified, but overruled by management
- Arianne V - Sensor failure as a result of reused hardware/software
-> Where do draw the line between an upgrade, and a new feature