Risk management - Monte Carlo simulation output
Monte Carlo simulation output
After approximately 1000 computer simulations we can get a histogram of the range of ‘total costs’.
Let us call each of these calculations a ‘simulation’. After all we are ‘simulating’ mathematically what might happen in practice.
Looking more closely at the histogram we can see:
- The axes are ‘frequency’ and ‘total cost’ in this example. This means that for each simulation we arrive at a ‘total cost’. The number of ‘total costs’ giving the same result are plotted in the frequency column.
- We know the MINIMUM ‘total cost’ as this is just all of the activity ‘minima’ in their ranges added together which could have a value, say = £50,000. Similarly, we know the MAXIMUM ‘total cost’ as it is just all of the activity ‘maxima’ in their ranges added together which could have a value, say = £250,000.
- The histogram shown, actually shows the frequency of ‘total costs’ over a ‘range’ of values. In this case there are 20 intervals each one being £10,000.
So, in this case, we can see that in the first interval (£50,000 to £60,000), there were no occurrences in any of the simulations. As we move further along the ‘total cost’ axis we get 50 occurrences for the interval £120,000 to £130,000 and further on we get 30 occurrences in the interval £160,000 to £170,000. At the far end, we see may be 1 or 2 occurrences for the final interval of £240,000 to £250,000.
The histogram confirms what we knew previously that the probability of all activities being at the LOW end or the HIGH end of the range is extremely small.
We can use the histogram to consider risk [see Monte Carlo distribution].