Reworked route restrictions (aka "bar-sum rule")

What?

Replace the foundation of the so-called bar-sum rules with something that is not a moving target like the traffic numbers we use right now. This will likely require new attributes to be added to and researched for our airport data because using the technical size of an airport (like runway length, size category) often isn’t representative of an airport’s actual relevance when it comes to preventing slot blocking (which route restrictions are effectively about).

Why?

As a reminder: The bar-sum rule is called this because it is based on the “sum of the bars of a route’s two endpoints”, with those bars currently being based on the total passenger demand at these airports. As passenger demand constantly changes, the rule can start or stop applying to a given route-aircraft-combination at any point in time.

Naturally, this is far from ideal and we should seek a more robust mechanism, especially now that location-based demand uses a completely different definition of “demand” such that an airport’s “bars” and its actual size often do not correlate at all anymore.

When?

Technically anytime.

2 Likes

Maybe link the restrictions to the slot usage?

  • If “slots < 50%”, you need “average plane size +1”.
  • If “slots < 20%”, you need “average plane size +2”.

By this, an airport would start to fill itself. Lets say, DASH is size 2, CRJ is size 3, A319 is size 4, A321 is size 5… at some point with a fleet of only DASH, my airport hits 50%, so i need to step up for new routes and use bigger planes?

Just as a small idea with no linkage to real numbers. Maybe even link the required plane size to the hour? So even busy airports can be target of small planes if they do not hit the waves?

Maybe link with the “size” of the airport, so small airports have a “cap” on the max required plane size? (e.g. an airport will never require planes larger than “airport size”, that is set by a combination of “runways, terminals, passenger demand,…”)

What shoud have to be considered:

  • Handling existing route assignments: How handle a change in planes if you have “old routes” in place..

    • Allow only changes that meet the requirements?
    • Allow a change only to a larger plane although it does not fullfill the requirements?
    • Allow a change to the same plane type?
    • Allow no change at all?
  • Slot blocking

  • Allow at least a small amount of small(er) aircraft, especially for new players…

Another thinking:

based on slot usage, you apply a “gaussian curve” to allowed aircraft. Like:

  • 10% size >=1
  • 20% size >=2
  • 40% size >=3
  • 20% size >=4
  • 10% size >=5

So in my example, if i fill up to 50% of slots in an hour, my next plane would need to be size 3. Larger airports could have a “higher” distribution (allowing only smaller amouts of smaller aircraft), smaller airports lower requirements (e.g. Sylt would only require “737/A320” for the top 20%).

This would fix the problem of replacing old aircraft, their “slot” for small aircraft would still exist, and also for new players and smal(er) aircraft…

(after thinking about both, i would prefer the second aproach as it would be easier and can be adjusted to airport sizes)

Interesting ideas!

The root issue I see with both is that whoever comes first can effectively do slot-blocking up to the permitted level without any consequences, which is exactly the problem this feature is supposed to address. It’s not so much about people scheduling the occasional turbo-prop to a large airports…it’s players getting as many of the cheapest (= smallest) planes as possible as early as possible to then block slots before anyone else can use them.

Yeah, for slot blocking, you would have to think different.

#1: enforce the distribution from the beginning. Even for small airports, you need a “bigger bird” at some point until you can continue to use smaller. The bigger the airport, the higher the demand for bigger planes. So for "early blocking”, i cannot spam 10 LETs, but maybe i can have

  • 1 LET (first plane not limited?)
  • next needed is 3x something in the range of a Dash
  • then you need 1x something like 737/A320
  • then you can continue with Dash again
  • at some later point you can maybe have a LET again…

But still, this could lead to a situation where YOU get the 737, but another player is quicker and fills all “smaller” slots before you can!

#2: Make the distribution (gaussian curve) apply for each player individually! By this, each player is limited by himself. The airport size determines which “distribution of size” would be best. Each player can still have small planes to an airport. With limited holdings, its a first step. Maybe make it apply for all companies of a player (or start at holding level) so you dont spam subs to fill an airport.

Starting players who do not have a lot of starting money still could start with a rather small plane, even on “high demand routes” (FRA-LHR), although the bigger players are forced to increase their plane sizes by time. With the new customer types, new players could focus on a small, but maybe profitable group.

Been following this one for several days & I’ve read this through several times & I keep coming to the same thoughts that I just want to highlight. I think finding the right balance for this topic is crucial for finding a system which provides the balance we’re looking for while not compromising too far down the route of game mechanic.

The difficulty lies in the fact that outside of gate sizes & the infrastructure servicing available at each airport, only really equates to the operational costs for the airline. Otherwise if an airline has the finances to buy into whatever slots they’re after then really they can put on the route pretty much any aircraft within the performance window of that airport.

I don’t think the current system is inherently bad but perhaps some tweaks could be introduced which could improve the baseline concept. What I would say is that I don’t think it would beneficial to introduce a set of rules that can become too convoluted & confusing.

Perhaps another way of looking at things could be to focus on the behaviour patterns when it comes to slot blocking. From what I’ve seen, it appears that the worst offending of slot blocking tends to occur on a very quick timeline, use the same kind of meta aircraft & sometimes also is combined with harsh price dumping. Perhaps this could be detectable via a script?

This could expand the moderation options & compliment whatever changes are made for this system. Also understanding the behaviours you want to remove might also help to create a solution that would serve best to restrict that kind of play.

We could always do something similar to what we did for “transfer capability”: That flag used to be based on airport size…we just took the values according to the given airport size on day X and made it a dedicated attribute which is now managed independently. We could do the same with the “bars”, basing them on today’s demand values and then manually set them going forward. Wouldn’t improve the solution per se, but would remove the fluctuation issue and thereby buy us some time to come up with a better system.

The whole point of the system is to no have to “moderate” slot blocking :slight_smile:

1 Like

Buying some time could be worth considering since we know there’s a fairly big list on the roadmap. & anything that can help provide a bit of extra flexibility when it comes to development certainly isn’t a bad thing :wink: Im also thinking that once a few more of these different systems progress onto testing, I reckon there’s a good chance that a solution to route restrictions might even present itself.

1 Like

I think the current system does it’s job pretty well. Before it was introduced we used to have weekly cases of slot blocking. Basically a player would schedule 100s of LETs (or similar) between massive airports thereby blocking the market for himself until he could afford to actually develop it with proper planes. The route restrictions changed that over night.

As long as any current flights are grandfathered when demand changes I don’t see a reason to change the current system.

1 Like

I know it cuts off a bit of realism - but the current rule works for me. Years ago back I set up in CDG and a competitor set up with a huge fleet of Bn2 islanders. He was outed on the server and forums, and I think eventually banned by mods - but it was fustrating.

2 Likes

That’s how it used to be. It was horrible.