Pipe Risk Assessment Tool
You can use the tool to prioritize your water distribution and transmission pipes for renewal projects. This prioritization will identify the set of pipes most at risk for failure and with the greatest cost-based consequences for inclusion in your capital improvement program (CIP). The tool is for analysis of pipe failure risk as part of a utility’s asset management program. It is not intended to replace enterprise asset management systems or to be used to manage data on all utility pipes.
Non-Login Site: www.erams.com/prst
To utilize the tool, a shapefile with the following attributes will be needed:
This exact naming system is not required. The tool does give the user the ability to select which layers correspond to which attribute once uploaded.
To upload the file:
- Click the “Map” tab from the left dashboard.
- Select “Geospatial Layers” > “Add Layer” > “Add Shapefile”.
Once your Shapefile is uploaded, you can zoom to extent to see your file. An example of the pipe system for Colorado State University is shown below:
Next select the layer with risk assessment information. For example:
Choose the “Year Installed Field” and adjust the “ Likelihood Index” as a function of age for analysis. Example is shown below:
Choose the field for “Materials” and adjust “Modification Factor” as needed:
Choose the field for “Service” and adjust “Likelihood Index”:
Choose the “Pipe Breaks” field and adjust “Likelihood Index”. Set the “# of pipe breaks”, “Pipe Length”, and “Breaking Status” as well.
Adjust the Overall Likelihood Index based on priorities:
Select fields for cost estimation and adjust the “Consequence Index” with estimated cost:
Select the Desired Combination Level (1, 2 or 3) and then evaluate the model:
Likelihood Index Description
Four major items of information are required to estimate an overall likelihood index. The first is the relationship of pipe age to an estimated likelihood index function. A likelihood index function provides an intermediate calculation step to compute the likelihood of failure. All likelihood index functions have a range of 0 to 1.
In the “Age” Tab, the user may enter up to nine values of age and the associated likelihood index function values.
A table of potential likelihood index modification factors as a function of pipe material is provided under the “Materials” tab. The names or abbreviations of the pipe material in this table must exactly match those in the pipe inventory tab. The default modification factors are typically equal to one; however, if experience indicates that a certain pipe material has a higher failure rate compared to other types of material the factors can be increased to numbers greater than one and the likelihood index will be multiplied by this factor.
The table containing the relationship of service conditions to an estimated likelihood index function is provided in the “Service Conditions” tab.
The table containing the relationship of the number of breaks to an estimated likelihood index function is provided in the “Pipe Breaks” tab. To normalize this data, the relationship is based on a break ratio which is defined as the actual number of breaks for a pipe segment as compared to the national average of expected breaks for that pipe segment. The range of this ratio is between 0 and 2.
The Overall Likelihood Index Calculation is based on a combination of the likelihood index values for Pipe Age, Service Levels and Pipe Breaks. The relative importance factors for these three items is entered in the table beginning in the “Likelihood Index” tab. These are ratio values and therefore a value of 2 is twice as important as a value of 1. An estimated consequences of failure rating is computed as a function of pipe diameter, location and repair difficulty. The user defines relative importance factors for these items in the table.
As an option, the user can estimate a consequence cost in dollars as a function of the consequences rating. The user provides this information in the table beginning in “Costs” tab.
Overall likelihood index of failure = f (age, breaks, service)
To compute the overall likelihood index, the tool uses a weighted average combination of the likelihood indices for age, breaks and service:
Overall likelihood index of failure = w1*LI(age) + w2*LI(breaks) + w3*LI(service)
The tool allows the user to specify the relative importance of the indices which in turn are used to compute the weights (w1, w2, w3). Weights are computed by dividing the individual relative importance values by the sum of all the relative importance values. Figure G.5 shows an example of the table of relative importance factors that are specified by the user (Cells B70:D73). For example, the relative importance factors for Level 3 (1,2,1) will produce weights of 0.25, 0.5, and 0.25.
The tool provides three options or levels to compute an overall likelihood index, using combinations of age, breaks and service conditions. The three different methods of overall likelihood index computation provided are defined below based upon the relative importance factors shown in Figure G.5:
Level 1—based 100% on age (w1 = 1.0)
Level 2—based 50% on age and 50% on breaks (w1 = 0.5 and w2 = 0.5)
Level 3—based 25% on age, 50% on breaks and 25% on service condition (w1 = 0.25 ; w2 = 0.5 ; w3 = 0.25).
Information Required For Each Pipe Segment
PipeID is a unique identifier given to each pipe segment in the data base. The PipeID is used as the primary identifier of each record and must be provided. If a utility does not have a unique identifier they might use a set of sequential numbers to serve as the PipeID.
Length (ft) is required for each pipe segment. It is used in the calculation of the expected number of pipe breaks for the age of the pipe.
Material is required for each pipe segment. This may be entered as a name (cast iron pipe) or as initials (CIP); however the entries for the material must be consistent throughout the database. This will allow the user to filter subsets of pipe segments based upon the type of material the pipe is constructed of. If the pipe material is unknown, a name of “Unknown” might be used so that those pipes can be filtered into subsets as desired.
Installation (year) is required for each pipe segment. This is used in the calculation of pipe age (current year – installation year). If the installation year is not known, it might be approximated by the installation years for other pipe segments in the same area.
Diameter (in) is required for each pipe segment. This is used in the calculation of the consequences score. It is also useful for filtering subsets of pipes.
No. Breaks is required for each pipe segment. This is the number of breaks a pipe segment has experienced and it is used in the calculation normalized break ratio, that is, the ratio of the actual number of pipe breaks divided by the expected number of pipe breaks. If a pipe segment has had no breaks then a 0 should be entered.
Breaking Status Factor is required for each pipe segment. The value of this factor would typically be set to a value of 1. However, if a pipe segment has experienced more than 1 break in a short period of time (for example, within the past 5 years) this might be an indication that the likelihood of breaks is increasing for this pipe segment. The likelihood index computed for the number of pipe breaks based upon age and pipe segment length is multiplied by the breaking status factor. For example, if the utility estimates that a pipe segment is breaking at twice the expected rate, then the breaking status factor should be set equal to 2.
Service Factor is required for each pipe segment. Service factors will have values of 1, 2 or 3, corresponding to a word scale of light, medium and heavy. This is an indicator variable that measures threats such as traffic load, high pressure zone, corrosive soils and any other service condition that might cause the pipe to fail more quickly than expected.
Location Factor is required for each pipe segment. Location factors will have values of 1, 2 or 3, corresponding to a word scale of remote, average and difficult. For example a remote location (factor = 1) would be used where a location is easy to get to and less likely to cause collateral damage. By contrast a difficult location (factor = 3) would be used where a location is hard to get to and more likely to cause collateral damage and liability payments.
Repair Difficulty Factor is required for each pipe segment. Repair difficulty factors will have values of 1, 2 or 3, corresponding to a word scale of easy, medium and difficult. For example an easy repair (factor = 1) would be used where the access, depth of the pipe segment, and other aspects of the site are amenable to an easy repair. A difficult repair (factor = 3) would be used where access, depth of the pipe segment, unique conditions such as a stream crossing, or a combination of issues would create a difficult repair situation.
Priority is an optional input. This input was suggested to allow users to identify pipe segments with priority for repair or replacement. An example might be pipe segments under a roadway that is scheduled to be rebuilt. This input is used only as a means to filter subsets of pipes. Priorities can be identified using either numbers of an appropriate word scale. This field can be left blank if desired.