Loadfactor issue

[kcclieou]

Airline details: Delta Air Lines [Private world #737; ID 22903]
Route IDs involved: 1355, 1671
Type of bug: suspected loadfactor bug

In private world #737, I discovered an anomaly in loadfactor calculation, exemplified by the following two instances:

1. Aircraft ID 173, an Airbus A319, is deployed on the route DTW - CLE (route ID 1355), and since extra hours are available I decided to increase the frequency from one roundtrip flight each day to two, anticipating that this would bring me extra revenue (despite its being less optimal).
The following screenshots display what happens before and after the increase in frequency.





With the same fare with which I can attain a 100% loadfactor by flying one roundtrip each day, I can only attain a 32% loadfactor if I fly two roundtrips each day. This sounds blatantly ridiculous - by flying twice each day, fewer passengers are willing to fly with my airline even though I offer the same fare - I mean, it is reasonable that I cannot achieve a 100% loadfactor on the route if I fly it twice each day with the optimal fare for a daily flight, but it is not reasonable at all that my loadfactor should drop to anything below 50%.

2. As part of my long-term expansion plan I changed the equipment on the DTW - BNA route (current route ID 1671) from a Boeing 717-200BGW to an Airbus A319. The optimal fare is $360 on a B717:


When I upgrade the equipment to an A319 and apply the same fare, the loadfactor drops to 66%:


Now a B717 seats 117 passengers while an A319 seats 145. This would probably mean that I should be able to seat 117 passengers on the A319 if I apply the optimal fare ($360) for the B717, resulting in a 80.7% loadfactor. But again, it turns out that an equipment upgrade without a change in fare ends up reducing the number of people willing to fly on the route.

The two instances sound so strange that I would suspect that there is a bug in the script for loadfactor calculation. Can someone please look into the problem, or explain to me the rationale behind such anomalies if this is not a bug at all?

[MrOrange]

To keep it simple, it's supply and demand.

In the case of the 1-frequency to 2-frequency A319 route, you may have been very close to the total demand when you supplied passengers with a 1-frequency route. Demand doesn't change if you don't change your prices, so if you create a 2-frequency route for the same price, demand %-wise decreased because of the increase in supply. As this calculation is not so much a linear process, your loadfactor could drop under 50%.

In the second case, same thing applies. First, you offered 117 seats for price X, now you offer 145 seats for price X. Again, as loadfactor calculation is not so much linear, your loadfactor could drop under the 80% you calculated.

That said, we are looking into the loadfactor calculations to see if we can make improvements. Your post will be looked at again at that time, I'm sure.

[kcclieou]

Demand doesn't change if you don't change your prices, so if you create a 2-frequency route for the same price, demand %-wise decreased because of the increase in supply. As this calculation is not so much a linear process, your loadfactor could drop under 50%.

Ahem. So that's the point. Demand percentage-wise decreases, but the actual demand appears to decrease as well simply because I'm putting extra capacity on the route. Would you mind explaining a little bit of the 'nonlinearity' of the loadfactor calculation?

[MrOrange]

Logically, you'd be right. Sadly, the loadfactor doesn't seem to be so logical. It doesn't decrease as you figured it would, it decreases more strongly (non-linear, as in a quadratic curve). While demand, following logic, appears to change, my strong guess is it's just the loadfactor calculation that doesn't follow economics. We'd have to look into the formula to be sure demand doesn't change, but as far as I know it doesn't. Most likely we won't figure that out until the entire loadfactor calculation is done, mostly because we don't really have the manpower to suitably adapt the current formula.

[Air Elbonia]

I'll look into it a bit closer, there's a number of factors that play into demand but yes, in theory if 117 want to pay $350 for a ticket given a similar enough aircraft in terms of speed and size [to some extent. Theres a point where small aircraft are impacted negatively for their diminutive stature, as in real life].

It is, for better or worse, at the moment a supply driven economy, which is where some of the screwiness happens.

I do tweak periodically in the background the demand formula & expenses. Never anything drastic, as most people have differing views as to what the proper demand should be on a route. 

Random factoid: the create, edit, and close route pages are accessed about 14 times as often as the main index page for AM.