0

I have the value of production waste (m3/day) and also the consumption (kwh/day) of some locations. You can see below that I have total information for locations 1 to 10. However, I don't have a consumption value for units 11 to 15. Therefore, I would like to know if it is possible to find these values, based on the consumption values I already have. It's important to note that they don't have to be exact values, but rather an estimate of what the value could be.

   Locations  Production Consumption
1          1  239.936000      467.36
2          2  422.183520       795.2
3          3    5.863376       176.2
4          4   23.993600      467.36
5          5  406.091680       738.5
6          6  143.961600     2226.36
7          7   42.348704      107.13
8          8   61.679680      198.63
9          9   12.956544       210.3
10        10  182.058268     1198.96
11        11 6168.500000            
12        12  714.593000            
13        13  268.545000            
14        14  175.200000            
15        15  227.577500
Carlo Soares
  • 3
  • 2
    Welcome to MSE! You're asking us to find predictive modeling for you... This is a so broad topic, that you are unlikely to get a response to your question. I suggest that you have a first look at regression analysis to analyze by yourself what kind of model can fit your problem. – mathcounterexamples.net Jul 20 '22 at 20:14

1 Answers1

0

Ploting the available data, it seems natural to fit the data with a function of the form $g(p) = \alpha + \beta \sqrt p$. Using the least squares method, you see that best fit is $g(p)=0.7533 + 46.1266 \sqrt p$. In the graphic below you can see the data points (blue) and the fit (red).

However this was quite arbitrary... Maybe there are aspects of this specific situation that advise a different approach.

enter image description here

PierreCarre
  • 20,974
  • 1
  • 18
  • 34