Add new scenarios
Add new variable
Spreadsheets: Equations Ranges

Breakdown and crossover analysis

Investigation of effect of assumptions on comparison of alternatives, e.g. within cost-benefit analysis

Quickstart

Load managed aquifer recharge demo - explore the effect of changes in assumptions on the financial value of managed aquifer recharge ('basin' and 'injection' scenarios) compared to storage of water in dams ('base').

Load diet water footprint demo - explore the effect of changes in food group footprints on the total water footprint of recommended ('RD') and meat-reduced ('A0) diets.

Introduction

Some analyses, e.g. cost-benefit analysis, involve comparing the value of two or more alternatives. The results depend on estimates of many variables, which are often uncertain. It can be useful to check how the conclusion might change with different assumptions about the estimates of these variables: is it possible that a 'cross-over point' will occur, such that the alternative that has the greater value will change depending on the assumptions we make?

This tool steps through this analysis, allowing you to change assumptions used along the way.

    Initial analysis:

  1. Define the equations used to calculate the value of each alternative
  2. Define the minimum and maximum values of variables that we want to explore
  3. Breakdown analysis:

  4. Break-down the initial result obtained for each alternative, showing where it comes from
  5. Cross-over point analysis:

  6. Identify cross-over points where values of single variables are changed
  7. Identify cross-over points where values of pairs of variables are changed
  8. Identify cross-over points where values of many variables are changed

Initial analysis

Several options are available: The analysis can also be saved and wiped clean:

How much does a single variable need to change to reach a crossover point?

For each variable separately, we aim to identify the value of that variable at which two alternatives have the same output, i.e. where a cross-over point occurs. The cross-over points are ranked and coloured by their level of concern/level of comfort. (Show more)

Levels of comfort and concern depend on the best guess, minimum and maximum values. They are justified and can be edited in the pane on the right, for the variable selected in the table.



Why is the best guess value selected:


What influences the bounds:
  • Is this crossover point of concern? Why/why not?
  • What should the bounds be?
  • If there is no crossover point, is this expected? why?

  • What direction of change is expected? Does it match the model results?


    Given this crossover point exists, can we conclude that any of the options should be ruled out? Why?
  • What further analysis might show the crossover point is not of concern?
  • What can we do to avoid this crossover point?
  • Colours are used to highlight levels of concern:
    Add Delete
    Add Delete