Suppose you desire to go on vacation next year. You may like to estimate whether the amount that you will be able to save during next year would be sufficient to do so. You could first create an influence diagram to show how the key factors that determine the amount of estimate your (new or additional) Savings next year i.e. from your Salary and your Living Expenses:
The influence diagram shows the relevant variables and directionality of the relationship between these. It indicates that Salary and Living Expenses are inputs (i.e. they have no arrows pointing into them, and are also coloured blue for additional clarity), and (new) Savings is the output (no arrows pointing away from it, and coloured green for additional clarity).
Although it is not stated explicitly on the diagram, it is clear that to calculate (new) Savings one needs to subtract Living Expenses from Salary. This can be captured by using a direct numerical calculation in Excel, resulting in a simple model to estimate the Savings:
The estimate is that your new Savings would be $2000; you can then consider whether this is sufficient and then decide whether to go on vacation or not.
This model is a simplification of the underlying real-life situation in many ways. For example, it does not take into account that you may already have other savings or sources of funds that could be used to partially pay for the vacation. Also, it does not take into account any possibilities to adjust your expenditure, nor the detailed breakdown of the expenditure, nor the possibility to increase savings by reducing expenditure on some of the less essential items. It also does not take into account whether these items (or the salary) would changed during the next year, nor by how much, and nor the timing of any such changes within the year.
Finally, the context also assumes that the only decision available is “vacation or not?”. There may be many other (and possibly better) ways to use these saving for consumption or investment.
In other words, there are many aspects that could (and perhaps should) be considered further. However, the saying “every model is wrong, some are useful” is pertinent: Models are used to improve (and ideally, optimise) decisions and to support decision processes.