In the case of Bennett Solutions, if the support assembly does not fit, do we need to build the tree into a recursive model of considering buying another new part from Miami supplier or buying another from ARC, etc. can we just end that branch of the tree taking into account the cost of returning the product and the reduced refund price? A: There is no set answer, so make whatever assumption you think is reasonable, state it clearly, and proceed with it.
Q: Is the cost of rebooking per passenger another decision tree? What we are trying to figure out is the % of load (occupied seats) and % of passengers looking to rebook same day etc. that are to be used as fixed for calculations or as uncertainities. A: Try to make as few assumptions as possible.
As a general guideline, when the case gives you enough data (e. g. , historical numbers, estimates, etc), such that you can compute a number directly, try to do that, instead of introducing new elements in the decision tree.
Q: Is it possible to fly an empty plane from Santiago to Miami in order to avoid a cancelled Miami – Santiago flight? A: Carefully read the notes at the end of the case — one of them should stipulate something about this. Q: How long does it take to make the actual repair to the system once the parts arrive in Santiago? A: Make an assumption that you consider easonable, state it clearly, and proceed from there… Q: Consider the sentence: “However, since this is a current initiative, there is only a 35% to 50% probability that support assemblies of different fleets will fit perfectly and hence are interchangeable. There is no way to know in advance, this has to be checked on the aircraft when the component is installed”. Does the above apply only to the part supplied by Bennett or it applies to the other suppliers? A: It is safe to assume that it only applies to Bennett Cargo Sales as it refers only to the support assembly.