9 replies to this topic

[WillaLotz]

Hi everyone! This is my first posting so please bear with me.

I would like to know how the teams determine their optimum race strategy. I would guess that they use some sort of simulation model they developed and using this try and predict how the race would unfold to a certain degree so that they can build their race strategy around this.

Does anyone know to what degree this is true and how it works?

Thanx
Willa

[red300zx99]

I would guess that they were trying to perdict the outcome of the race, and chosing whatever strategy gives them the best chance to...........one could only guess. Some simple matrix game theory is all I know, but I would think it's the same idea's, only more F1(complex). The number of different inputs that could be used in the model, such as your pace, that of your competitors, where the the slow cars are on track, fuel, tires, maybe even alittle of a reliablity factor even. Maybe Im crazy but I think all of these things and more are helpful in determining what to do and when. But I bet the people who work on that stuff are some rather interesting folks.

[WillaLotz]

I'm currently working on such a simulation model for a final year project in Industrial Engineering. Here are some of the things I am taking into consideration:

Some Factors to take into consideration when creating the simulation model:
• Tyre wear –
• The frequency of need for fresh tyres (see FIA rules and regulations, subjected to only 1 tyre per qualifying and race) – only one set of tyres per race!!!!!!!!!!!!
• Penalty a driver carries when tyres are worn
• Other tyre manufactures information (only for 2005,subjected to change to only one tyre manufacturer after 2005)
• Tyre punctures which result in a unscheduled pit stop and a new tyre

• Fuel consumption –
• Weight penalty for carrying fuel
• The specific fuel consumption for all cars on the starting grid – estimations which can be calculated using previous races’ pit stops information
• Frequency of need for fuel
• Distance fuel load will take the car in a race

• Time it takes passing traffic –
• Ease to pass traffic – over taking opportunities (which will differ from track to track , example Monaco its nearly impossible)
• Time spend in traffic or to pass traffic

• Pit lane -
• Length
• Pit lane speed limit
• How the pit exit and entry is designed – the ease of entering and exiting the pit

• Weather forecast or conditions -
• Air temp
• Track temp
• Wind forecast
• Rain forecast
• Mist forecast
• Air Pressure (only per track) - High altitude = thinner air = l less aero drag + less down force...

• Race track information –
• Number of laps in the race
• Length of circuit
• Race distance
• Track surface : tyre wear, break wear
• Track Altitude (change in altitude): influence the performance of the cars
• Race is stopped if longer then 2 hours

• Pit stops -
• Time for tire change
• Refuelling rate – maximum of 12.1litres/second
• Quantity of tires available to a car – according to FIA rules and regulations

• Qualifying strategy and results –
• First Qualifying time
• Second Qualifying time
• Aggregate time to determine grid position

• Current track conditions -
• Due
• Track dampness or lack of dampness
• Rain
• Flag conditions on track (green, white, yellow….)
• Oil
• Debris
• Crazy fans

• Extra race laps in a race (need to take into consideration for fuel purposes)
• Warm up, before race
• Reconnaissance lap – from the pits to the racing grid
• Formation lap fuel – lap before race starts
• Cool down lap – at the end of the race

• Opposition team information –
• Number of opposition teams
• Number of opposition cars (maximum of 24 cars aloud on the grid – FIA )
• Opposition cars top speed
• Opposition cars lap times
• Determining opposition teams pit stop strategy - guessing or based on historical trends
• Race numbers used during the race

• Information regarding the different tyres available to the team –
• wet
• intermediate
• dry soft
• dry hard

• To determine team under investigation’s strategy -
• Race starting fuel
• Grid position of all cars on grid including the team under investigation
• State of tyres which will be used at the start of the race
• Projected weather conditions

• Car retirements –
• Crashes
• Failures
• Disqualifications
• Scale of points – how points are awarde

The what if questions which need to be answered by the simulation model:
Different scenarios to be simulated • Safety car (accident, bad track conditions – too wet ) –
• Duration of safety car period
• Location of incident requiring safety car
• Flags on track – could be present without the presence of a safety car

• Weather changes -
• Air temp,
• Track temp,
• Wind
• Starting to rain or stops to rain

• Penalties -
• Drive trough penalty
• A ten second time penalty
• Drop of 10 grid positions – qualifying or from previous race penalty – engine
• Rules concerning serving penalty ( according to the FIA ) – FIA must inform team not later than 5 minutes after incident has occurred and the penalty must be served within 3 laps after notification
• Time penalties
• False starts
• Driver stalls on the starting grid
• Special starting procedures in the event of rain
• Race stopped – what procedures apply (safety car)
•

• Changing strategy in mid race due to –
• Traffic
• Cant pass a lead car (must do it in pit stop)
• Tyre wear (graining of tyres)
• Opposition teams’ car failure, made a mistake or got served a penalty

• Unscheduled pit stop -
• Reacting to other teams strategy’s/pit stops
• Damage on the car (loss of front wing)
• Safety checks on car after an incident
• Change in strategy due to the above mentioned

• Racing incidents which may influence strategy ( fuel etcetera )
• Tyre punctures
• False starts
• Collision
• Starting from the pit lane – complying to FIA rules and regulations
• Fans on track
• Debris on track
• Oil on track
• Traffic voids that has developed – want to take advantage of this if near scheduled pit stop

• Changing tire Pressures during the pit stop

[Greg Locock]

You might want to think more about which is the most /robust/ strategy. Not necessarily the optimum, but the one one that gives you the best chance of finishing in a near minimum time given a reasonable mixture of possible events.

Rumour has it that the truth is that the car setup stuff is moot, the cars are optimised for qualifying.

[jcbc3]

I seem to remember an interview with Ron Dennis some years ago where he stated that when they (McLaren) made their strategies, they computed what would be the fastest way of driving the 300km. 1, 2 or 3 stops and such. They did NOT take into account what-if scenarios, because it only muddled the picture. That was also why you very rarely would see McLaren drivers on different strategies. And IF they were on different strategies it would be because they had qualified a substantial number of places away from each other.

But this may of course be redundant information. Maybe they DO utilise the different what-if scenarios now.

[WillaLotz]

jcbc3, I hear what you are saying, but what happens if say a safety car comes on track at lap 17 of a 54 lap race and you are currently on a one stopper?. Do you bring in both your drivers at that particular moment and thus change your strategy to two or maybe even 3 stopper or stay out and keep to the one stopper? You only have at most 60 seconds to make that decision, but if a simulation was done say the previous day and you did do a run which anticipated the safety car you would be able to know what the effect of such a strategy change might be.

Thus with such a simulation model you would be able to have a rough idea of what to do and what the out come might be. Which gives you an advantage. I also realise that the human factor(the Rhos Brawns' of the formula 1 world) can never be done away with - no model can ever be perfect and that model can not have the experience and feeling a human have, but it can be of good guidance.

The problem I am having at this stage is not if such a model is needed or used but how deep in detail do one have to go into in order to get the model to represent the real world problem.

I cant imagine that a team would not make use of an state of the art simulation model to look at different scenarios of what might happen on track during the race and how to respond to them.

[jcbc3]

sorry, maybe I didn't make myself clear.

I was only talking about the basic race starting strategy. Take Australia as an example. McLaren had Kimi and JPM on the same strategy and fel loads as they qualified 9 and 10. Suddenly this was thrown out the window as Kimi stalled and had to start from the pit lane.

Did McLaren already have a plan B for that happening? Maybe. But when yu start all the way back there are all kinds of permutations of how a race can unfold. Meaning that they should have a new strategy for all positions and time differences between drivers. Which they can't.

So they play it 'by ear' most of the time. Which is where Ferrari has a major advantage. Because they have Ross Brawn. I doubt you'll find much disagreement here about him being the master tactician. E.g. last year at the French GP. Here's a story from Grandprix.com:


...The TV cameras seemed to be fatally drawn to the blue cars although now and then they followed Michael Schumacher's Ferrari after he had got ahead of Alonso using a smart slash-and-burn four-stop strategy. The Renault was quick and because overtaking is so tough at Magny-Cours might have stayed ahead if Ferrari had stuck to a conventional strategy but such is the self-confidence in this team that when Race Strategist Luca Baldesserri came up with the idea of doing four-stops, Michael was happy to give it a try.

"We had nothing to lose," said Ross Brawn. "There was no risk because we were stuck in second and it would have been difficult to overtake. We needed to run in free air and to be able to use the car."....


So yes, they do plan ahead but maybe not to such a degree as you seem to think. Maybe there's a job waiting for you. Sometimes you see a race on the telly and think, that you could do a better job than they do in the teams. I seem to remember that in "the old days" when there was some kind of emergency you took in your lead car and let your second car circulate for one extra lap. Often costing the hapless driver oceans of time. Now the conventional wisdom is to pull in both your cars and let the second wait in line for service. Frustrating but ultimately much less time wasting. It's so obvious when you think of it, but still it took a while to catch on.

[Lukin]

I think that the teams would have their pre-race strategy set, and during the race continually update the what-if scenario's. If a SC comes out, do we stop? If so, how much. Where will it put us in regard to traffic. I cant imagine it would be too hard, all you need is the live timing, which each team has.

[CARVER]

If you have a car that is a lot faster than all the other cars you have a lot more options to get a winning strategy (which is not always an optimal strategy, time it takes to finish the race). The Ferrari case with 4 pitstops I think it would not have been possible if the Ferrari had not been a dominating car. Getting a clear track (free air) is very important. I think there are a few main points that are important:

- don't get stuck behind a slower car (that can't keep up with the pace of your competition). So you need to do a good qualifying and start. You cannot qualify too light, because you will lose more time in the pits and you might come out in traffic while being heavy and those behind you running light which might allow them to overtake you and the one that is being overtaken usually loses a lot of time with that.
- If you are stuck behind your competition (and you're faster) try to stay out longer (unless it will get you in traffic, in that case pit at the same time and do a faster stop because you still carry more fuel and so need less fuel than your opponent)

So a lot of this depends on what the competition does. This can be a problem when you're in a close fight with more than one opponent.

I don't know if teams do this, but having two cars they can also come up with a strategy that uses both cars. One light to get in front of faster cars and then block them by running slower than it can and drive very defensive. The other car heavy so it actually runs it own pace which is the same pace as the others are running behind his teammate. This way you destroy the chances of both cars getting a good result (say: 9th and 10th), but you increase the chance of one car getting a great result (say: 6th). I don't know if this is allowed and it is not particularly fair sportsmanship, but it is a way for the teams in midfield to score some points.

[WillaLotz]

Originally posted by Lukin
I think that the teams would have their pre-race strategy set, and during the race continually update the what-if scenario's.

I'm just wondering why there seems to be a lack of information regarding how teams go about determining their strategies. Right, I know that they wont give away secrets but it almost feel that there is so little out there regarding this issue. If you wont to know anything else about F1 you are almost certain to get good reading material on the particular subject you wish to educate yourself on, but the subject of how/who/what they do to determine their strategies seems to be not documented at all. Maybe it is not interesting to other people ? I on the other hand think that there could be said allot on this issue and it makes for good debate.

So if anyone know some detail on the issue raised, please let me know