Perform a sensitivity analysis

Perform a sensitivity analysis to identify the inputs whose variation have the most impact on your key outputs and demonstrate the effect of changing the standard deviation of the inputs.

What is a sensitivity analysis?

Sensitivity analysis allows you to assess the results and identify the inputs whose variation have the most impact on your key outputs. Use this method along with your process knowledge to identify the inputs that can be adjusted to make improvements. Companion displays a graph that shows the effect of changing the input standard deviation on the percent of output that is out of spec.

Look at the graph for these patterns:
  • To increase the capability by reducing the variation in Y, look for inputs with sloped lines. This pattern indicates that a change in the variation in the input will likely reduce the variation in the output, resulting in an increase in capability. The graph shows that changes to the inputs indicated with red and blue lines are candidates for improvement efforts. For example, in construction project scenario introduced in Add a Monte Carlo simulation, you can reduce the percent of projects that run longer than 30 days if you are able decrease the variability of these inputs.
  • To identify inputs that have little or no effect on the variation in Y, look for inputs with a flat line. For inputs with a flat line, you can ease the requirements (tolerances) without adversely affecting the performance, which will save you time and money. The graph shows that changes in the variation of the purple and green effects have little influence on the percent of projects that exceed 30 days. Therefore, you may want to evaluate the impact of relaxing the requirements on your business.

Sensitivity analysis often follows parameter optimization, which focuses on finding optimal settings for the inputs. For more information, go to Perform a parameter optimization.

Perform a sensitivity analysis

Use the sensitivity analysis to evaluate the effects of the input variation on the output variation.

  1. Choose Simulation > Sensitivity Analysis.
  2. If you have more than one output, a drop-down list appears so that you can choose the output that you want to examine.
  3. Examine the graph.
    • Look for inputs that have sloped lines. Consider the relationship between changes to the standard deviation of the input and the % out of spec. For example, you may want to generate results that show the effect of reducing the standard deviation of the Execution phase by 10%.
    • Look for inputs with flat lines. These inputs have little effect on the variability so you may be able to reduce the tolerances. You can bring this information back to your engineering team for consideration. For example, you may want to evaluate the benefits of relaxing the requirements for the Delivery and Execution phases.
    Tip

    If you prefer to evaluate the standard deviation rather than the % out of spec, click Standard Deviation on the y-axis label. (% Out of Spec is only available when you have specification limits.)

    If the lines are close together, you can isolate an input line by choosing it from the graph legend making it easier to select and evaluate changes.

Examine the results

Companion displays the results of the sensitivity analysis, assumptions, and guidance for next steps.

  1. View the results in the workspace. The parameter optimization for the construction projects example indicated that around 8% of the projects will extend past 30 days. Following the sensitivity, you can now expect around 7% of the projects to extend past 30 days.
    Note

    Each time you repeat the simulation, the results will vary because the simulation is based on randomly selected values for the inputs.

  2. Under Assumptions, compare the new settings to the previous settings and confirm that the new settings are feasible for your application.
  3. You can do any of the following tasks:
    • To view DPMO (Defects Per Million Opportunities), observed performance, summary statistics, percentiles, and model assumptions, click More Results.
    • To view information about next steps, hold the mouse pointer on .
    • (Optional) Perform another sensitivity analysis. Typically, you make changes to one input at a time so you may want to evaluate changes to other inputs.
  4. Repeat the sensitivity analysis with new settings until you are satisfied with the results. You could also repeat the parameter optimization. For more information, go to Perform a parameter optimization.

Switch the view

Each simulation has two parts: the model and the results. Use the ribbon buttons to toggle between the two views.

Edit the model

After you analyze the results, you may want to return to the model and change inputs or outputs, and then rerun it. This allows you to test a number of "what if" scenarios allowing you to gain insight into the behavior of your system and make better decisions.

  1. To change the model, choose Simulation > Edit Model. By default, Companion overwrites the existing values in the current model. To preserve the current settings and work in a copy instead, choose Simulation > Edit Model > Copy Current Model.
  2. Choose Simulation > Simulate.
By using this site you agree to the use of cookies for analytics and personalized content.  Read our policy