The Cohort Model Analysis can be used to predict the service life of groups of facilities. This method will group pipes with similar likeness (e.g., material types, pipe diameter, or soil types) and then create a deterioration model for each group.
Cohort Models may be found beneath the Sensitivity Analysis option in the Operation Center - Analysis tab. Right-click to create a new Cohort Model. 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.
You may also right-click on created Cohort Models to be able to quickly view previously generated reports and set your analysis as a map display.
Cohort Model Settings
The first step is to set the basic Cohort Model parameters. The first window should resemble the example below. In this window, you set the x-axis by setting the From Year and To Year fields, enter the Target Year for which detailed data will be produced for, and specifies the Facility Scope.
After clicking Next, the second window contains more detailed parameters for you to enter.
Type of Model - For Cohort Models, you have the option of selecting Herz, Weibull, Non-homogeneous Poisson Process (NHPP), or Homogeneous Poisson Process (HPP) model equations. For more information on the differences between the four, see the Model Equations page. Innovye's general recommendation is that the Herz model should give better predictions since it is water industry specific, but if this model does not converge, the Weibull model, which is more general, should be used. NHPP and HPP model equations are more advanced and can also be used as desired.
Grouping by - Because this is a Cohort Model, you have different options for choosing how you wish to group your facilities.
- Facility Attribute: You simply select a field from your facility model to group by. After you specify a field you must also specify...
- Input Data: This is the failure data which must be fed into the Cohort Model. You can either generate and specify your own failure curves with the Life Expectancies option, or utilize a previously created Failure Definition. The former option is a more manual process using previously generated curves, while the latter is a more automated InfoAsset Planner process.
- Sensitivity Analysis: With this option, you will be selecting the grouping and the failure data at the same time. You select a Sensitivity Analysis (which contains an associated Failure Definition) and then select a categorical variable to group your cohorts.
Cohort Model Results
Once the parameters are set, click Run to view the Cohort model results. The results will open in the Dashboard and the following options will be available to you: The graph on the right-side shows the different cohorts as different colors.
Result Type - You may use the drop-down menu to access five different graph categories calculated in the Cohort Model.
- Failure Probability Density: This measures the likelihood that a given cohort, or the average pipe in that cohort, will fail in any given year. In the example below, the average CIPP pipe (yellow curve) has about a 2.4% chance of failing in 2040.
- Cumulative Failure Probability: Displays the likelihood that the average pipe in a given cohort has already failed by any given year.
- Failure Rate: Shows the number of times the average pipe in a given cohort might fail in any given year. This is slightly different than the Failure Probability Density since it is a decimal and not a percentage and it can exceed one and predict more than one failure in a given year.
- Survival Probability: The likelihood that the average pipe in a given cohort has not experienced a failure in a given year.
- Residual Life Expectancy: This is the only bar chart in the set. It shows the life expectancy of the average pipe in a given cohort. You can adjust this graph by editing the Failure Threshold value on the left.
By Year/By Age - You can adjust the display of the x-axis with these radio buttons. By Year will match the constraining values set in the parameters.
Optimistic/Pessimistic - You can add Optimistic and Pessimistic curves to your graph. These cannot be generated if you are using manually set Life Expectancy curves.
Life Expectancy Setting - These options are only available when viewing the Residual Life Expectancy bar graph. You can edit what percentage the failure threshold should be set at in order to edit your graphs.
You may also save your models/graphs, zoom, pan, and toggle to the report tab as desired.