InfoAsset Planner's Sensitivity Analysis helps determine which parameters are important in determining whether a facility has failed or not.  Sensitivity Analyses look at the Failed vs. Non-Failed facilities grouped in the Failure Definition and then evaluate a number of variables specified by you to determine whether there is any pattern between the variables and the failure rate of facilities.


The Sensitivity Analysis tool may be found in the Operation Center, within the Analysis tab, under the Deterioration Model drop-down menu.  You can right-click the Sensitivity Analysis option to create a new analysis.  Like many saved items in InfoAsset Planner, the ID must have no spaces or special characters, while the description has no character restrictions and is much easier to change later.


Sensitivity Analyses must contain an earlier created Failure Definition which is specified in the Failure Definition ID drop-down menu.  Next, you may specify a Threshold of variable non-significance probability.  This variable determines how strict the Sensitivity Analysis should be in determining whether a variable is significant to pipe failure or not.  The smaller the value, the more likely the Sensitivity Analysis will exclude the variable as significant. Users should determine this value carefully; too low and InfoAsset Planner won't be able to find any relationships between the variables and won't be useful, too high and InfoAsset Planner will be too lenient and may produce less reliable Cohort and Regression Models.




After clicking Next, you can begin selecting their variables for to run through the analysis.  To add a new variable, click  and reference the appropriate data.  Tabular data can come from the Facility Data attribute table, other data internal to the InfoAsset Planner project database, or external data which has been added to the Table of Contents.  The basic requirement is that is must be able to be joined to the facility data via Facility ID.  You can edit previously created variables by using the button.  Make sure to check the tick boxes to include the created variables in your Sensitivity Analysis.



After running, InfoAsset Planner should produce a table in the IA Dashboard similar to the one below.  In this example, Material, Diameter, Slope, Peak Score, and Risk were all determined to be significant variables which could help predict pipe failure.  No relationship or pattern was found for the 5 Year Rehab Action, Length, Depth, and Mean Score variables so they were not selected.  The Category Detail field shows how a categorical variable may be grouped.  In the example below, CIP, PVC, and CO pipes were grouped together because InfoAsset Planner found no statistical differences between these variable values.  So instead of grouping these pipes strictly by Material so that there are eight separate cohorts, the Sensitivity Analysis found only four significantly different categories.  The Baseline Value field is simply the average value or the mode depending on the type of variable selected.