The life cycle cost analysis calculations presented herein were completed using a combination of two widely available software programs, MicroSoft Excel (deterministic) and Palisade's @Risk (probabilistic) as suggested by the Federal Highway Administration (5). This combination allows both the deterministic and probabilistic approaches to be analyzed. Before describing the scenarios investigated, an overview of the general features of each is described along with the assumptions.

Scenario overview

  • Several features were common to each analysis. These are described below:
  • A 40-year analysis period was selected based on FHWA recommendations (5).
  • Each major rehabilitation activity triggered a lane rental cost calculated as a function of the production rate assumed for that traffic level/facility type.
  • Routine (preventive) maintenance may be applied between major rehabilitation activities depending on the agency.
  • Salvage values were calculated as a prorated percentage of the expected life of the rehabilitation.
  • All costs were converted to present worth terms to compare asphalt rubber and non-asphalt rubber alternatives.

In addition, several assumptions and simplifications were necessary. These are listed below:

  1. 1) Maintenance was applied as indicated by the agency. Once triggered, maintenance costs occur until the next major rehabilitation activity.
  2. 2) User delay costs were approximated using the lane rental costs. The authors recognize that more accurate costs could be determined if actual average daily traffic (ADT) were known and delays were computed; however, this was beyond the scope of this project.

Several input were consistent among all different scenarios. These included:

*Four percent was used for all deterministic runs; when variability was considered either 2.5, 4.0, or 5.5 was used for a given calculation.
Variable   Input Values
Discount rate, %   2.5, 4.0, 5.5
Analysis period, yrs   40
Lane rental costs, $/lane-mile/day   1,000; 5,000; 10,000
representing low, medium, and high traffic levels, respectively
Project length, mi
+ City/county projects
+ State DOT projects		   5
10
Production rates, lane-miles/day    
+ City/county projects
+ State DOT projects		   2
3

 

The maintenance and rehabilitation strategies used as well as the expected lives and costs varied as described below.

Scenarios investigated
The scenarios investigated were presented in Table 1. For each agency, several rehabilitation and maintenance strategies were evaluated. The expected lives and costs for all rehabilitation and maintenance strategies used in the analysis are given in Tables 2 and 3.

Results
In the deterministic approach, variability of the inputs is not considered. For the scenarios evaluated, the net present worth values are summarized in Tables 4-6. It should be noted net savings result from the use of AR in most cases for the inputs used in the analysis as indicated by ratio of costs > 1.0.

For those situations where the asphalt rubber alternate was not cost effective (e.g., ratios < 1.0), the following were determined:

  1. 1) Maricopa County (Alternate B vs D). For the asphalt rubber to be cost effective, the average expected life would have to be 11 years (instead of 10) to yield a ratio > 1.0.
  2. 2) City of Mission Viejo (Alternate K vs L). The estimated life of the AR alternate must be increased by 1 year (from 20 to 21) for the AR alternate to be cost effective.

It should be emphasized that all of the estimated lives are best estimates provided by the agencies. Any change in estimated life can have a significant effect on the LCCA.

For the probabilistic analysis, the input variables were selected randomly within the ranges given for all inputs except the following:

  • Analysis period, fixed at 40 years
  • Lane rental costs, fixed as determined by traffic level
  • Project length, fixed as determined by traffic level
  • Production rates, fixed as determined by traffic level

Figure 3 illustrates the approach used and the interpretation of the results of these calculations are shown schematically in Figure 4. In this example, alternate A would be more cost effective 77 (100 - 23) percent of the time.

Figure 5 provides the results for the City of Phoenix. If the percentile where the costs for each alternate within a scenario are equal (0 savings) is determined, then the data in Figure 5 suggests that asphalt rubber is more cost effective than the conventional alternative in many applications considered (see Table 7). As shown in Table 7, the following observations are made:

  1. 1) City of Phoenix. The results indicate the AR alternate to be cost effective in all applications except when comparing alternates A vs C (high and low traffic). These results indicate it is more cost effective to use microsurfacings or slurry seals prior to placing the AR overlay. Deferring major rehabilitation to later will always be more cost effective.