FIND MARKET PRICE and SUPPLY AT PRICE return erroneous results
Posted: Fri Apr 18, 2025 8:58 pm
Hello,
I have been testing using FIND MARKET PRICE and SUPPLY AT PRICE to estimate electricity dispatch, assuming normal distributions for different power plant dispatch costs.
I was doing some validation of the results and found that the results from Vensim are erroneous. You can validate this against what should be computed using normal mathematical operations, for instance in Excel.
In my example, I assume 6 power plants: coal, NGCC, NGCT, nuclear, solar and wind.
Dispatch costs are 25, 35, 55, 10, 0, and 0. I used a normalized StDev of 0.25, so multiply that by each of the average costs to find the standard deviation.
Total capacity for each are 200, 300, 200, 500, 100 and 100, and all are assumed to be fully available in every hour.
Using the the FIND MARKET PRICE function, i get a market price of $9.1872 in the first period when demand is set to 400.
This yields supply of 7.96552 for coal in this period using SUPPLY AT PRICE
In Excel, I computed the expected total output at that price using the same parameters using the NORM.DIST function. At that price, it only results in coal dispatch of 1.14. Across all resources, the total dispatch in Excel is 387.9759 when it should be 400. This is a rather large error on the order of 3%, which of course compounds over many hours and resource types.
Am I doing something wrong here or is the function(s) not working correctly? I am on Vensim 9.2.4 but can easily verify this on a later version. I've uploaded my Excel workbook and an example model.
I have been testing using FIND MARKET PRICE and SUPPLY AT PRICE to estimate electricity dispatch, assuming normal distributions for different power plant dispatch costs.
I was doing some validation of the results and found that the results from Vensim are erroneous. You can validate this against what should be computed using normal mathematical operations, for instance in Excel.
In my example, I assume 6 power plants: coal, NGCC, NGCT, nuclear, solar and wind.
Dispatch costs are 25, 35, 55, 10, 0, and 0. I used a normalized StDev of 0.25, so multiply that by each of the average costs to find the standard deviation.
Total capacity for each are 200, 300, 200, 500, 100 and 100, and all are assumed to be fully available in every hour.
Using the the FIND MARKET PRICE function, i get a market price of $9.1872 in the first period when demand is set to 400.
This yields supply of 7.96552 for coal in this period using SUPPLY AT PRICE
In Excel, I computed the expected total output at that price using the same parameters using the NORM.DIST function. At that price, it only results in coal dispatch of 1.14. Across all resources, the total dispatch in Excel is 387.9759 when it should be 400. This is a rather large error on the order of 3%, which of course compounds over many hours and resource types.
Am I doing something wrong here or is the function(s) not working correctly? I am on Vensim 9.2.4 but can easily verify this on a later version. I've uploaded my Excel workbook and an example model.