Gazzz's tutorials : Scheduling

[gazzz0x2z]

( disclaimer as Hostingpics is about to close, I had to actually upload the images in the forum, so they now appear at the end of the post, sorry for the inconvenience)

Scheduling Basics

Hi everyone. Having mentored countless beginners, and having seen the same questions - and mistakes - countless times. So what is a good schedule?

A good schedule, basically, is using the plane to its maximum. A plane makes money in the air, not in the ground. The more blue you have on your schedule, the better. But how to do that?
The other rules to remember are :
_you are penalized strongly in terms of Load Factor for taking off or landing between 0000 and 0455. You are penalized lightly, bit still, for taking off or landing between 2300 and 2355 and 0500 and 0555.
_you are penalized strongly in terms of Load Factor if you take off with less than one hour of difference between two flights on the same route - but only if you have opposition(don't worry, it comes quickly - someone who starts 3 daily flights at 0600 for the same destination has identified himself as a target). This sime span get lower for destinations with very heavy demand. Around 2000 demand, 30 minutes are usually enough between two flights. NB : start time of your opponents is always irrelevant. Just their number of flights is
_you are penalized in terms of delays and cancellations if you don't allow any margin between flights. 737-700 are able to do a turn-around in 40 minutes, but it's safer to aim for 70 minutes.

The first thing to do is to know you fleet. With a 737-700, for example, I know I can cumulate 2200NM per days if I fly twice during the day. Or 1550NM for three flights. Or 900NM for four flights.
Other jet single-aisle aircraft will have similar values. Regional jets are different, as they are usually slower, but with a better turnaround time, so they'll shine on short distances. You can even setup 5 daily very short flights with some of them. I did it with CRJs.

Which means, from Orly, with my 737-700, I can setup 4 daily flights to Strasbourg(217NM), or 3 daily flights to Figari(518NM), or 2 daily flights to Kiev-Borispol(1108 NM). Of course, I can mix destinations, with one flight to Kiev and two to Strasbourg.
Let's have a look at the simple method.
(first image, down the post, the scheduling image)
It's on the upper line, on this image. The F-BABE plane. 3 different destinations. Orly-Cork(456NM), 0520-0605/0720-1000, Orly-Prague(471NM) 1110-1255/1405-1600, Orly-Porto(648NM) 1710-1830/1940-2245

Several remarks on this schedule, which is good, IMHO, but not perfect. I had constraints when setting this up(like my other lines, limited slots, and the fact that Orly closes at 2255). Additionally, it's a port from a 737-300 schedule, which makes it even worse.
1)I'm flying 1575NM. That's just a little bit too much, and explains some of the imperfections. A certain level of imperfections is acceptable. Don't take off at 0200, make a minimum turnaround(40 minutes for the 737-700), and land at 0400
2)The flight to Cork is the most punished. Taking off at 0520 is already punishing, far more than 0550, for example. My load factor is higher on the back leg, and it's not random. Ideally, I'd set up a lower price on the leg to Cork.
3)The turnaround time is important away, but also at home between two flights. This I did respect between each flight. It's important too, as it's very tempting to respect the full turnaround time in the schedule, but to stick 2 flights with only th minimum turnaround time between them. I could make the flight to Cork take off much later(with therefore a much better Load factor on the first leg), but I'd be punished in terms of delays and cancellations. Delays are damaging your image, cancellations are costing you a lot of money.


OK, this is good, but there are juicy destinations from Orly for 737-700 that are much further, and it's a shame if I fly only one a day. What to do? Well, in some cases, you can set up red eyes flights

A red eye flight, basically, is "take off late in the evening, fly to the East, and land early in the morning". Depending on the relative position of your base, the other leg will be before in the evening or later in the morning. From Orly, most destinations suitable for that are in the Middle east, so you'll be back in the morning. For Boston, suitable destinations are in the West part of the USA, so the red eyes flight is the leg back.

Let's have a look at the F-BABK here. It's flying to Boston from Orly during the night. It works because demand is not too high. If demand were above 220, it would be suicide to set up a 737. But as it's more in the 150, it works, and the plane is full and makes money. It's long for a red eyes flight : 1240-1510/1620-0600. The secret is to add a complement destinations 6 times a week; in this case, Newcastle(408NM), 0710-0750/0855-1135. Not a very good destination, there is a lot of opposition over there, but it's a secondary destination. The plane makes its money on the Boston route, the 12 daily k$ of the Newcastle route are just a bonus.

The important part here is : 6 times a week. Not 7. Because of the A-Check. And the airplane flies 19 hours per day in average. Good use of capital. Not my best plane(very long range 737 routes are never gushing money), but still a good one. Don't start a company with that kind of flights, though. You'll notice some 65 minutes of turnaround, here also, I packed as much as I could.

Note also that the Distance a plane can daily fly is bigger. You have to make a few tries with your favourite fleet group to know it well, and know what distance you can afford daily. With a 737-700, you can notice you're not far from 3400NM for 2 flights


There is a third type of "simple" scheduling, that is not as efficient as the first too(day-only or red-eyes), which I'll call "sleep-outside". It's inferior, but sometimes, you don't have the choice. Either because you're short on slots around 0600(a common problem for whoever plays in London City, where I learned the technique), or when you already fly 8 daily flights to a destinations, and still have some demand to fill.

The second problem is common in Orly, as it's a domestic airport, with a few key destinations having a lot of demand. So, you already have flights starting at 0600, 0700, and don't want to begin at 0800 because you'd lose too much time. Unless... Unless you make a short flight late in the evening, 6 times a week, with a huge turnaround, around 400 minutes, and you fly back to your base in the morning.

Which allows me to stack one more flight to Strasbourg. Not on saturdays, as I need a night for the A-Check, yet as saturdays have lower demand, it's not a big deal.

(second image down the post)

As you can see, I'm pushing a little bit : 2150-2255/0500-0610. You guessed it, my load factor is lower on the way back from Strasbourg. You'll have strategic choices to make between "more flights", and "better flights"

Anyways, with those three methods, Day-Only, Red-Eyes, and Sleep-outside, you should be able to successfully use all planes up to the "large" size.

For the very large ones, and distances beyond 3400NM, you need advanced techniques.

For reference : my other tutorials :
costs

[gazzz0x2z]

Scheduling Advanced

Using Very large(like the A330 in the following examples) aircraft to link far away destinations is a rewarding, yet challenging action in Airwaysim. Very large aircraft cots insane amounts of money to buy, or, even worse, to lease. Their overhead and staff costs are prohibitive. Which means they'd better always fly and make money, if you want to survive a few years, or, even better, make good money out of them.

There are three techniques used by the experimented players. Two are similar, the 1-plane-7-days technique, and the 1-plane-6-days technique, and the third one is completely different : the 7-planes_7-days technique.

The image below shows you all the details of a 1-plane-7-days flight, as well as a 7-planes-7 days technique

The lowest plane in my list, the F-SAEG(tough to see, that's the eight and last in the list) is doing a 1-Plane-7 days technique, for one single destinations, Chicago, in this case.
Saturday : 1350-1625/1855-1040
Sunday : A-Check 1210-1710, then flight 1750-2025/2255-1440
Monday : 1710-1945/2215-1400
Tuesday :1630-1905/2135-1320
Wednesday : 1550-1825/2055-1240
Thursday : 1510-1745/2015-1200
Friday : 1430-1705/1935-1120

This one is nearly perfect, as 3599NM is the perfect distance for this exercice. Everything takes off or lands between 0600 and 2255, all the available time is exploited, and the 150 minutes turnaround for the minimum delay risk of 1%. One is not always that lucky.

The 1-Plane-6-days-schedule is exactly the same, but for longer distances : you skip one day(usually the saturday/sunday night), and setup 6 flights, and can fill up distances beyond 4400NM.

This method is efficient, but when the distance is not perfect(not all juicy airports are exactly 3599NM away from Orly, after all), you may cringe seeing some of your flying time unused. Here comes the 7-planes-7-days schedule. On my screenshot, it's the 7 other planes doing just one of them.

Yes, the 7 planes schedule requires 7 planes. I know, one gets surprised by it, sometimes.... The idea is to begin on monday, fill as much as you can with different destinations, with a A-check, and finish the next monday, leaving just enough time for the turnaround. Easier said than done, especially in smaller airports like Orly, where juicy destinations are not in big numbers.

So, how does it work? Well, I've got 7 A330s, from F-SACT to F-SACZ. The F-SACT starts on monday, the F-SACU on tuesday, up to the F-SACZ that starts on sunday. And they all to the same, with one day of offset.

the F-SACT flies so :
Monday : A-Check 0000-0500, Saint Denis 0540-2100/2330-0900
Tuesday : Le Lamentin 1130-1555/1825/0835
Wednesday : Pointe A Pitre : 1105-1525/1755-0755
Thursday : Seoul 1030-0600/0830-1305
Friday : Dakar 1535-2030/2300-0555
Saturday : Taipei 1530-0610
Sunday : back from Taipei, 0840-1520

That one is far from my best(had no other juicy destinations when I made it), as it's not fully used. You can see two holes, one small between Dakar and Taipei, and a bigger one after the flight to Taipei. Still, it's good enough to make money.



Now, the one billion dolar question : which is the best? 7-planes or 1-plane? Each one has its own qualities and drawbacks. Most of the best players are choosing the 7-Planes schedule, as it has the huge advantage of making a better use of your costlier planes, and allowing more destinations. Taipei, in my example, is just little too far for a 1-plane-6-days schedule.
1-plane schedules, while more limited, allow other cool things : a more dedicated seating, especially, that allows you to tailor your aircraft for maximized income. Taipei has 7% buisness demand and 1% first demand, while Dakar has 3%/0%. Which means expensive seats are empty when flying to Dakar. 1-plane schedules also allow quicker moves. If you want to take a destination right now, and not let opposition the time to realize that you're going there, the 7-planes schedule is too slow to set up(time for the 7 planes to arrive). OTOH, 7-planes schedule will reduce your costs and increase your capability to cover a lot of destinations, and with it your capability to resist an attack from an opponent.

I'm mixing both, and it's not random. Both are efficient, and have their uses.

[gazzz0x2z]

Of course, feel free to ask questions, correct potential mistakes, or add additional knowledge on the topic.

[Andre090904]

Simply perfectly explained. That will surely help some people here.

Maybe you should add the little detail that the departure time of your competition is not relevant. I have seen many people "attacking me" on my departure times thinking they will steal passengers like that. But that's not the case...in this regard it may be worth mentioning that the total demand on a route will be equally shared between the number of flights (given the RI, CI, price, seating etc. is similar). Not many people seem to understand this.

[gazzz0x2z]

Thanks Andre. I added your first point. Your second one belongs to route choice, another tutorial I want to make(about route mapping and route choice) sooner or later.

[qunow]

- Is it necessary to aim for that 1% delay chance, with for instance leaving full 3h20m between each legs for A380?
- Is it necessary to leave some room for flight depart before and after A-check other than just that 5 hours maintenance time? How about for B-check scheduling?
- So even if ten other players all flown same route daily departing the airport at 2255, it won't affect my load factor no matter i depart at 2255 or 1055?

[Andre090904]

1) Not necessary, but recommended
2) No I would say, because maintenance has no delay in the simulation and it is being doing regardless if the flight right before the maintenance has been delayed. Correct me if I'm wrong
3) Exactly

[gazzz0x2z]

2) No I would say, because maintenance has no delay in the simulation and it is being doing regardless if the flight right before the maintenance has been delayed. Correct me if I'm wrong

That's perfectly right AFAIK. The 1-Plane-7-Days schedule example I am giving Has exactly 5 hours for the A-Check(when I say the distance is perfect...)

[LRoDgy]

Is there an easy way to calculate the best range for a 1-plane-7-day schedule depending on the specific aircraft?

[qunow]

Is there an easy way to calculate the best range for a 1-plane-7-day schedule depending on the specific aircraft?

168 hours, minus 5 hours A check, divide by 7, then minus (2*turnaround time), multiply the time with aircraft speed kts figure. the resultant figures in nm would roughly be the optimal distance with some nm in excess due to the time needed for landing/take off. try and calculate a bit for the exact distance.

[LRoDgy]

Thanks, that makes a lot of sense, should probably have been able to figure that one out by myself 

[paddk989]

Just like to thank you Gazzz for spending the time to write the tutorials. They are a necessity to become competent player in the game. Thank you

[paddk989]

I have been able to compile a 7 day schedule for 7 aircraft without a problem thank you. I am unable to do the 7 day schedule for 1 plane. Can you confirm which boxes need to be checked to get the correct route screen to compile the schedule. Thanks

[gazzz0x2z]

I have been able to compile a 7 day schedule for 7 aircraft without a problem thank you. I am unable to do the 7 day schedule for 1 plane. Can you confirm which boxes need to be checked to get the correct route screen to compile the schedule. Thanks

Well, I'm using multi-tabbing intensively when doing that. A 7/Day-1/Plane schedule is just 7 seperate routes, with the same flight numbers, each on a different single day(so that you can keep the same number), and at a different time. Let's imagine I want to open another route like that(not a good idea, there is alot of opposition now, but just for the example) :

• Select Orly base by a click under the company name
• Click Route planning(from LFPO)
• Search and find KORD airport. Click on the link to reach the demand chart
• Right click on open new route, and choose "open in a new tab" in your browser(no, you can't do it with a smartphones. Real computers are far superior). Seven times.
• In each tab, confirm the routing as it is. It's the f***ing 21st century, tech stops are so 20th century.
• For the first tab, choose saturday only, flight at 1350, fleet group, turnaround if you didn't set up a standard. Don't validate yet.
• For the second tab, choose friday, flight at 1430, fleet group, etc.....
• For the seventh tab, choose sunday, flight at 1750.
• Check every tab before validating ; slots must be available, flights shall not overlap, turnaround times shall be respected between two flights, you have enough money for buying slots, etc...
• Click Confirm in each of the seven tabs. It buys the slots automagically, and bring you to the post-confirmation screen each time.
• Use one of the post-confirmation tabs to set up a marketing route, if needed(i.e. if it's a route opening).
• Use another of the post-confirmation windows to allocate the 7-in-1 lines to your aircraft(because you've got already an aircraft in order, arriving within 3 weeks, right?)
• Close the remaining 5 tabs. Voilà.

Note that multi-tabbing is also very useful for 7 days 7 planes schedule. I've got one tab per destination, when doing those. But, in this case, select "every day" and "Create each day as a separate route". And check, and re-check, and re-re-check everything before confirming, and don't buy those slots yet. Unless your 7 planes arrive within 3 weeks, but that's not common.

[paddk989]

Thanks Gazzz for your comprehensive explanation to my question. Thanks for your time.

[qunow]

Because of the vastly different speed for piston aircrafts and turboprop aircraft, when upgrading routes from piston, sometime it would need a total reshuffle on schedule to properly utilize those new faster aircrafts. In such case, how to properly schedule those aircrafts with minimal slot lost/minimal slot reacquistion fee in mind?

[paddk989]

You are 100% correct qunow. When I converted from the dc3 to the viscount, shortly after the game started, I lost at least 50% of the slots bought.

[gazzz0x2z]

Well, fleet change is another topic. In current GW2, when I replaced wayfarers(180 ktas) with CV240(240 ktas) (a move of debatable strategic quality, especially that early in a plane-starved game, but that's another topic...), I didn't lose much money. How that?

First, I moved routes to the new plane keeping the starting time. OK, you'll tell me, it's nice, but at the end, I've got a 90 minutes turnaround in Oslo, for a plane that is better used with 40/45 minutes of turnaround. Why bother with a quicker airplane yet? Because it's far cheaper to move the slot after. You'll pay only 10% to 30% of the slot cost when reducing to 40 minutes. For very small changes, it may even be free.

For smaller changes, it's more a matter of planning. When I play 737 classics, I tend to set up turnaround times at 65 minutes, so when I switch to NGs, it goes naturally to 70 minutes on average flights.

[qunow]

I see.. so if I am rescheduling most of the flights by hours off their original departure time then there aren't really much that can be done except for progressively accumulating changes...

[gazzz0x2z]

I see.. so if I am rescheduling most of the flights by hours off their original departure time then there aren't really much that can be done except for progressively accumulating changes...

And add a short flight, or move a short flight from another schedule, or reorganize the schedule to use less planes, etc..... It can be a rather interesting game of Tetris played with the scheduler, where you try to fill your airplanes while limiting the cost of slot changes.