How can you evaluate the sensitivity and robustness of your simulation model?

1 What is sensitivity analysis?

Sensitivity analysis is a method to examine how the output of a simulation model changes when you vary the input parameters or the model structure. Sensitivity analysis can help you identify the most influential factors, the range of uncertainty, and the potential scenarios for your simulation model. Sensitivity analysis can also help you validate your model by comparing the results with empirical data or expert opinions.

2 What is robustness analysis?

Robustness analysis is a method to evaluate how well a simulation model performs under different conditions or environments. Robustness analysis can help you assess the stability, reliability, and adaptability of your simulation model. Robustness analysis can also help you improve your model by finding the optimal or satisfactory solutions, the trade-offs, and the constraints for your simulation model.

3 How to conduct sensitivity analysis?

Sensitivity analysis can be conducted using different techniques, depending on the type and complexity of your simulation model. One-at-a-time (OAT) analysis involves changing one input parameter at a time and observing the effect on the output. This is simple and intuitive, but may miss the interactions among the input parameters. Global sensitivity analysis, on the other hand, changes all the input parameters simultaneously and observes the effect on the output, which is more comprehensive and realistic but may require more computational resources and statistical methods. Lastly, factorial design changes a subset of the input parameters according to a predefined plan and observes the effect on the output, which is more efficient and systematic but may require some prior knowledge and assumptions about the input parameters.

4 How to conduct robustness analysis?

Robustness analysis of a simulation model can be conducted by using various techniques depending on the purpose and criteria of the model. Scenario analysis is useful for exploring the implications and consequences of the model, while optimization analysis can help find optimal or near-optimal solutions. Multi-criteria analysis, on the other hand, is beneficial for evaluating trade-offs and compromises among conflicting criteria. All of these techniques are helpful in assessing the robustness of a simulation model.

5 How to report and communicate your analysis results?

After conducting sensitivity and robustness analysis, you need to report and communicate your analysis results in a clear and concise way. To do this, use graphical and numerical methods, such as scatter plots, box plots, histograms, tables, charts, or maps to show the distribution, variation, correlation, or comparison of your analysis results. Additionally, descriptive and inferential statistics can be used to interpret and explain your analysis results. For example, you can measure the central tendency, dispersion, relationship, or difference of your analysis results with mean, median, standard deviation, confidence interval, correlation coefficient, or significance test. Moreover, narrative and storytelling techniques can be used to convey and contextualize your analysis results. By following these tips, you can evaluate the sensitivity and robustness of your simulation model and improve your simulation modeling skills.

6 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?