This is one of the questions asked in my recent interview as an energy trading analyst. Company XT owns a power station capable of generating 100 megawatts (MW) of power.The station uses 10 gigajoules of gas to generate 1 megawatt-hour (MWh) of electricity.XT has managed to secure a gas contract which allows them to purchase gas at a price of $8/GJ.assuming that the power station can be turned on and off instantly, the questions are:1. If there is an unlimited supply of gas, what is the trigger price at which the power station should be turned on?2. If there are only 175 terajoules (TJ) of gas for the month, what is the new trigger price at which the power station should be turned on?Any smart people here? Cheers,Gary

Not sure, but using simple mathematics...1)$80002) For this part, the supply is just 175 TJ and the required amount is 720 TJ. Since as the supply is limited and the cost of producing is almost flat,either it can be8000*720/175Also as the supply reaches the limit, the price increases drastically(hockey shape for price demand curve)WOuld like to know if this is right???

You should have specified that $8000 for 100 Mw-hours of power (else it doesn't make sense).So, the price should be about 8 cents per kwh.Even though, it'll make sense to run it at 8 cents per kwh (most of the variable costs covered) ... no fixed costs get covered at this price.I don't see any reason, how the limit of gas availability should affect decision to run the plant or not ... as far as the price of gas is the same.------------------------------------What really is of concern to me is "Any smart people here? " ..... i thought they were here.All quants are supposed to be smart ... aren't they? -- they are fooling the unsmart part of the world by producing random numbers, and still get paid for producing them !I'd also like to see what, Smartness ("Applied to this question") produces

For part 2, if they trigger at 8000, their fixed cost wont be covered: Plant can produce 100MW, so it is waste of their resources if they do not run it full....I am also curious to know what smart answers are needed for this question...QuoteOriginally posted by: bhutesYou should have specified that $8000 for 100 Mw-hours of power (else it doesn't make sense).So, the price should be about 8 cents per kwh.Even though, it'll make sense to run it at 8 cents per kwh (most of the variable costs covered) ... no fixed costs get covered at this price.I don't see any reason, how the limit of gas availability should affect decision to run the plant or not ... as far as the price of gas is the same.------------------------------------What really is of concern to me is "Any smart people here? " ..... i thought they were here.All quants are supposed to be smart ... aren't they? -- they are fooling the unsmart part of the world by producing random numbers, and still get paid for producing them !I'd also like to see what, Smartness ("Applied to this question") produces

Sorry guys, I just realise that I missed some important information about this question. There is also a forecasting power price matrix for a month. I cannot remember the actual numbers, but it looks like this: Each row is a different period and each column is a different day. The numbers in each cell are the expected price for that period of that day in $/MWhr.Prices Day Period 1/4/05 2/4/05 3/4/05 4/4/05 5/4/05 6/4/05 7/4/05 8/4/05 9/4/05 … 30/4/051 68 67 60 84 74 72 55 65 63 … 632 67 68 66 71 73 70 61 63 65 … 633 69 68 63 70 68 90 70 62 64 … 624 67 67 56 69 57 100 62 63 61 … 625 67 65 56 68 57 110 74 62 61 … 596 66 61 56 69 57 100 80 62 59 … 617 68 56 56 69 55 104 70 60 58 … 628 69 55 56 69 55 106 79 79 59 … 629 69 55 56 69 55 100 58 69 58 … 6210 70 56 56 69 55 97 58 55 59 … 6211 69 61 64 69 55 97 61 58 59 … 6212 68 68 68 69 56 96 62 39 60 … 60Sorry about. And you are both right for first question. I also think the trigger price should be either:$8000 for 100MWh$80 for 1MWh8c for KMh,So we can have a look the second part. Cheers,Gary

For the second part, just arrange all the numbers in a descending order, and start allocating gas from the top.Where ever the gas run out or the price falls below $80/MWh, stop.This way, you'' be maximizing your revenue from the limited gas supplies (while on the other hand never getting into the zone of negative contribution).I don't see anything remotely smart in this answer... but don't see what I am missing.

The one bit you are missing is that those prices are expectations, not actuals. So you would have to continue carrying out that process with the remaining energy and the remaining half hours in the month, always using the actual for the next hour instead of the forecast.