INTRODUCTION Life Cycle Cost Analysis and its Problems Interest in life cycle cost analysis (LCCA) appears to be increasing among building owners and designers. Some of this attention may be attributed to a related and growing interest in “green” building technologies that rely in part on the durability and sustainability of building materials to minimize environmental impacts. The increasing economic sophistication required to finance modern construction projects may also be a contributing factor. Finally, new federal requirements for public construction may be stimulating the growing interest in LCCA. Regardless of specific causes, however, the growing interest in life cycle costing is clearly reflected in changing attitudes within the construction industry. According to a recent survey conducted by Building Design & Construction (“White Paper on Sustainability,” 2003), an overwhelming majority of the 70,000 building professionals surveyed agreed that building materials should be evaluated first and foremost on the basis of life cycle cost. Unfortunately, although many building professionals are increasingly interested in learning about the life cycle costs of key building components, few tools currently exist to help them compare the almost unlimited choices of competing building materials. In the case of commercial roofing systems, designers and owners must select from a wide variety of roofing membranes, each with an equally wide choice of design and component options and warranted service lives, varying from five to over 30 years. The sheer complexity of modern roof system choices obviously makes it very difficult to develop simple analytical tools. However, the lack of effective life cycle cost programs also may be linked to other factors. Problem 1: How Long Do Roofs Last? The first challenge to effective LCCA is the lack of consensus regarding the service life of modern commercial roofing systems. As an example, two of the most comprehensive studies of service life conducted in the roofing industry arrived at sharply different conclusions regarding the longevity of various roofing systems. Based on a survey of over 400 roofing contractors, Carl Cash (1997) concluded that traditional multipleply asphalt roofing systems could be expected to provide an average service life of 17.4 years, while EPDM roofing systems could be expected to provide an average service life of 14.1 years. In contrast to the Cash study, Schneider and Keenan (1997) surveyed over 20,000 actual roofing installations and concluded that the average service life of asphalt multiple-ply roofs was 13.6 years, while EPDM roofs provided an average service life of 17.7 years. Using the Cash study as a basis for LCCA may favor the use of multiple-ply asphalt roofing systems, while the data from the Schneider and Keenan study may favor single-ply systems. How can the concerned building owner reconcile such conflicting estimates of roof service life? First, some of this apparent conflict may be due to the use of a statistical average. Within the population of both the asphalt and single-ply roofs, there may be roof systems that perform much better than the average, perhaps well in excess of 20 years. In addition, these better-performing roof systems may have included a variety of design and component augmentations that contributed to extended service life. The published warranty offerings of roofing manufacturers may offer additional insight into the relationship between roof system design and roof longevity. Based on a review of the NRCA Low-Slope Roofing Materials Guide (2005), roofing systems appear to exhibit a consistent upgrading of components and application practices as the term of the warranty increases. As an example, almost all 20-year, multiple-ply, asphalt roofing systems require the use of high-strength, Type VI ply felts and redundant flashing details, while systems with lower warranty lengths allow the use of lower-strength felts and less redundant flashings. In a similar manner, the thickness of single-ply roofing membranes tends to increase as the warranty term increases (from 45 mils at 15 years; 60 mils at 20 years; and 90 mils at 30 years); while seaming and flashing requirements likewise increase as the warranty term is lengthened. Although the nominal warranty term and relative durability of roofing systems J A N U A RY 2007 I N T E R FA C E • 5 appear to be related, there are no studies currently available to clearly mesure this relationship. However, the use of nominal warranty length and the system augmentation associated with the warranty period may offer a reasonable starting point. Accordingly, this study will use typical manufacturer warranty length and associated system specifications as a basis for comparison. Problem 2: How Much Do Roofs Cost? The second hurdle to effective LCCA in roofing involves the actual costs associated with installing, maintaining, and replacing a roofing system. Surveys involving mock roofing bids conducted by the author over the past 20 years indicate the price of identically specified roofs may vary by as much as 25% to 75% across the United States. These price differences may be attributed to different labor and productivity rates as well as regional differences in roof system selection. Some of the variability in roof system costs can be addressed by combining surveys of actual contractor bids with rankorder price surveys, which emphasize the relative rather than the absolute difference between roofing systems. As an example, contractor price surveys conducted by the author indicate that a typical ballasted EPDM roofing system may vary in price from a little over $2.00 to more than $4.00 per square foot; while a similar, fully-adhered EPDM roof may vary between $2.50 and $6.00 per square foot. However, when these two systems are ranked by contractors in terms of relative cost, adhered systems tend to command a relatively consistent cost premium of 25% to 30% above a ballasted system. By asking the same contractors to rank a variety of roofing systems, a consistent cost differential can be obtained for comparison purposes, even though the actual costs may vary significantly from contractor to contractor. Accordingly, this study has employed cost estimates based on commonly available national construction data, but these costs will then be adjusted based on rank-order estimates from a survey of roofing contractors. Estimates of annual maintenance costs also may vary from survey to survey. Respondents to Schneider and Keenan’s 1997 survey reported annual maintenance costs running from $0.14 to $0.19 per square foot, while respondents to Cash’s 1997 survey identified these costs as varying from $0.09 to $0.15. Because even the highest of these annual estimates is relatively low in comparison to initial installation costs, the present study will apply the highest estimate of maintenance costs, or $0.20 per square foot, to all roof systems in this study. Problem 3: How Does One Compare Roof Systems With Different Service Lives? The final hurdle to effective LCCA is related to the common methodology used to calculate life cycle cost. Accurate life cycle costing requires that all anticipated costs be converted to present value. These costs should include the initial cost of installation, periodic maintenance expenses, and eventual removal and replacement costs: LCC = IC + MCPV + RCPV Where: LCC = Life cycle cost ($/sq. ft.) IC = Initial cost MCPV = Present value of all future maintenance costs RCPV = Present value of future removal and replacement costs This approach requires that all anticipated future costs be stated as the amount of money needed today to pay the future costs, given an anticipated discount rate or cost of money. In order to allow for a consistent comparison among alternative products, this present value must be calculated over a defined “study period.” Typically, this study period should coincide with the investment horizon of the owner. For example, if a building owner expects to occupy a building for the next 20 years, the study period for life cycle cost analysis should also be 20 years. Although the use of a common study period allows for an “apples-to-apples” comparison of different roofing systems, it may fail to account for several important economic issues. In the previous example, even if a building owner expects to occupy a building for 20 years, the same building owner will also expect to sell the building at the end of the 20-year period. If the roof on the building requires replacement after 20 years, the owner may end up paying for a new roof, either by agreeing to replace it prior to transfer to a new owner, or through a discount in the selling price. Conversely, if the roof on the building is considered to be suitable for an additional 20 years of 6 • IN T E R FA C E J A N U A RY 2007 use, the building owner will suffer little if any loss of value in the sale of the building. In either case, the arbitrary 20-year study period may misrepresent the actual costs of ownership experienced by the building owner. The use of an arbitrary study period also makes it difficult to effectively compare the value of roofing systems with different estimated service lives. As an example, the true value of a 15-year roof may be understated as compared to a 20-year roof if a 15- year study period is selected that provides no economic value for the additional five years the 20-year system offers. Conversely, the true value of the 20-year roof could be significantly overstated if a 20-year study period is selected that requires the complete replacement of the 15-year roof but then understates the long-term value of the new replacement roof. Both of these problems can be resolved by deducting the residual value of the roof from the life cycle cost calculation, but this may add unnecessary complexity to what started out to be a fairly simple statement of present value. A more effective alternative may involve the use of Equivalent Uniform Annual Cost (EUAC) in lieu of standard LCCA. Unlike LCCA, EUAC allows for the use of differing study periods by expressing costs as an annualized estimate of cash flow instead of a lump-sum estimate of present value: EUAC = (A/P, i, n) Where: EUAC = Equivalent uniform annual cost A/P = Annualized cash flow or payment ($/sq. ft.) i = annual interest rate (%) n = service life (years) In simpler terms, Equivalent Uniform Annual Cost is the “payment” required to fund the Life Cycle Cost over the service life. This “payment” is calculated using the same principles as mortgage financing. The Life Cycle Cost represents the “purchase price” and the Estimated Uniform Annual Cost represents the “mortgage payment” needed for a given interest rate to fully fund the Life Cycle Cost by the end of the stated service life. Because EUAC costs are stated as an annualized amount, it becomes possible to compare roof systems with different service lives. ROOF LIFE CYCLE ANALYSIS USING ESTIMATED UNIFORM ANNUAL COST (EUAC) Step 1: Identifying Alternatives and Timeframes Using the NRCA 2004-05 Low-slope Roofing Materials Guide (2005) as a reference, a wide selection of roofing specifications was identified based on warranty length. In addition, specifications incorporated all major categories of low-slope commercial roofing systems, including traditional asphalt, modified bitumen, EPDM, and thermoplastic systems. In all cases, these roofing designs increased in redundancy and augmentation as the warranty term increased. As an example, a typical 15-year EPDM specification may allow the use of a 45-mil membrane, while 20-year and 30-year designs require minimum 60- mil and 90-mil membranes, respectively. In a similar manner, a typical 15-year modified bitumen system may allow the use of a nonmodified fiberglass base sheet, while a typical 20-year system requires a modified asphalt base sheet. For the purposes of this study, the nominal warranty period was also designated to be the service life period for each roofing system. It is very likely that the actual ser- J A N U A RY 2007 I N T E R FA C E • 7 vice life may exceed the warranted service life, but the variation based on warranty length allows for relative comparison among the systems. The roof system specifications and warranty periods selected for the study are identified in Table 1. Step 2: Identifying and Calculating Costs INITIAL COST As mentioned in the introduction of this study, initial costs were developed using a two-pronged approach of 1) establishing initial costs using commonly available industry construction estimating data, and 2) modifying these initial costs using rankorder data derived from a survey of roofing contractors. Initial costs were established using Means Building Construction Cost Data 2005. This initial cost data were then adjusted in accordance with average rankings as identified in a survey of 50 commercial roofing contractors located throughout the United States who were asked to list the rank order of each system in terms of installed cost. The adjusted costs for each system as determined by this method are summarized in Table 2. REPLACEMENT COST In order to develop an effective life-cycle cost comparison, the cost for the eventual replacement of the roofing system must be determined. Unlike the initial roof installation, replacement cost will include both the equivalent cost of the initial installation as well as the tear-off and disposal costs of the original roof. Although costs for replacement after tear-off and disposal can be calculated using the original installation cost values, tear-off and disposal costs must be determined independently. One of the most comprehensive and consistent estimates of disposal costs can be found in a study of roofing durability conducted by Cash in 1997. Based on a survey of roofing contractors, Cash estimated that the removal and disposal costs for the types of roofing systems included in the present study varied between $0.83 and $0.98 per square foot. Because these costs appear to vary within a relatively narrow range that may have little significant impact on the outcome of the life cycle cost calculation, the current study assigned a uniform value of $1.00 per square foot for the removal and disposal costs of each system studied. In addition, this assigned removal and disposal cost was converted to present value in order to adjust for the timing of the replacement. In effect, this present value is Table 2: Roof system specification and initial cost. Infrared Roof Moisture Scans High resolution short-wave FLIR Thermacam for accurate scans on reflective surfaces. Survey results marked on the roof surface and verified. Full reports with CAD drawings on hard copy or via e-mail. Infrared Inspections, Inc. 1-800-543-2279 www.rci-mercury.com 8 • IN T E R FA C E J A N U A RY 2007 Table 1: Roof system specification and warranty/service life. equal to the amount of “cash on hand” that can grow at a given interest rate into the amount of “future cash needed” to fund the roof replacement. As an example, using an annual discount rate of 5%, the cash on hand or present value necessary to replace a roof in 15 years is equal to 56% of the future cash needed, while the present value of a roof that must be replaced after 20 years is equal to 46% of the future cash needed. (Please note that a discount rate of 5% was selected, as currently recommended by the Federal Energy Management Program. See Fuller & Rushing, 2005.) The present value or “cash on hand” replacement costs for each roof system specification are summarized in Table 3. MAINTENANCE COST As mentioned previously, annual maintenance costs were based on data from Schneider and Keenan (1997) and Cash (1997) that identified annual maintenance costs as varying between $0.09 and $0.19 per square foot. To simplify this current study, these costs were rounded up to $0.20 for every roofing system. In addition, the total maintenance cost for each roofing system was converted to present value (or “cash on hand”) by calculating the discounted cash flow of the annual costs for the warranty period. As an example, the cash on hand required to fund a $0.20 annual maintenance cost for a 15-year warranty period is $2.18, while the same present value for a 20-year warranty period is $2.63. These present value maintenance costs are summarized in Table 4. Step 3: Calculating Life Cycle Cost Once the present values of all initial, maintenance, and replacement costs have been established, the calculation of Life Cycle Cost is simply accomplished by combining these costs into a single amount. Table 5 summarizes all Life Cycle costs for all roofing systems identified in this study. Step 4: Calculating Equivalent Annual Uniform Cost The problem with a simple life cycle cost model becomes apparent in Table 5. The life cycle cost of many 15- and 20-year roofing systems is very similar, and in some cases, the life cycle cost of some 15-year systems is lower than the corresponding 20-year system. As an example, the present value cost of a 15-year modified bitumen system is only $7.81 per square foot, while the more durable and redundant 20-year modified bitumen system has a higher cost of $8.49 per square foot. The problem, of course, is that the 20- year system provides a longer service life than the 15-year system, and the value of this additional service life can only be evaluated by annualizing the costs associated with both systems. This can be accomplished by expressing the costs of both systems as an annual cash flow or “payment” for the expected life of each system. This annualization of life cycle costs is achieved using the Equivalent Uniform Annual Cost (EUAC) method, as previously identified in this article. Using the same 5% discount rate as in the LCCA calculation, the EUAC for each roofing system is summarized in Table 6 and graphically compared in Chart 1. (1) Present Value Adjustment Factor based on a 5% annual percentage rate applied for the warranty period or service life of the system. As an example, the initial present value or “up-front” funding necessary to cover annual maintenance costs for a 15-year service life requires 10.9 times the annual maintenance cost, while the “up-front” funding for a 20-year service life requires 13.15 times the annual maintenance cost. Table 4: Roof system, service life, and maintenance cost. 10 • I N T E R FA C E J A N U A RY 2007 (1) The Present Value Discount Factor is based on a 5% annual percentage rate applied for the length of the service life/warranty period. As an example, a 15-year service life requires 56% or 0.56 of the replacement cost “up front” as the present value of the future cost, while a 20-year service life requires only 46% or 0.46 of the cost “up front,” due to the longer time period to accumulate interest on the initial amount. Table 3: Roof system, service life, and replacement cost. Step 5: Analyzing Results: The EUAC Method of Life Cycle Costing LONG-TERM VALUE OF SPECIFICATION ENHANCEMENTS The EUAC method of life cycle costing may help to identify the real benefits inherent in roof systems that have been enhanced to extend service life. Based only on a comparison of the basic LCC of 15- and 20-year roofing systems in this study, the benefits of enhanced specification might be questioned because the LCC costs were so close. However, the EUAC cost method identifies that 20- and 30-year systems may hold an advantage more than sufficient to justify the additional up-front expense. As an example, the EUAC calculations indicate that the 20-year roofing systems in the study may offer long-term costs at 10% to 15% lower than their 15-year counterparts. In addition, the EUAC of the single 30-year system studied offers a cost savings of 12% beyond a similar 20-year system. The EUAC data in the present study appear to support the proposition advanced by many roof consultants that the investment in enhanced system design may provide a real economic return to the building owner. As perhaps best stated by Richard Boon (2001) in Roofing Contractor magazine, “…the higher up-front costs of premium roofing systems can be fully justified through long-term savings.” ECONOMIC SIMILARITY OF MAJOR ROOFING SYSTEMS The EUAC method also suggests that the major types of commercial roofing systems used throughout the United States today provide a very similar economic benefit. Although roofing industry professionals may hold widely divergent opinions regard- Table 5: Roof system life cycle cost (LCC). Note: Equivalent Uniform Annual Cost is the “payment” required to fund the Life Cycle Cost over a service life. This “payment” is calculated using the same principles as mortgage financing. The Life Cycle Cost represents the “purchase price” and the Estimated Uniform Annual Cost represents the “mortgage payment” needed for a given interest rate to fully fund the Life Cycle Cost by the end of the stated service life. Table 6: Roof system Equivalent Uniform Annual Cost (EUAC). Test your knowledge of coatings with the following questions, developed by Donald E. Bush Sr., RRC, FRCI, PE, chairman of the RRC Examination Development Subcommittee. 1. What are the five major reasons why galvanized steel is painted? 2. What is the most common reason for painting galvanized steel? 3. Which three basic components do protective coatings contain? 4. What are the three mechanisms by which coatings protect a substrate? 5. Which structural design features may contribute to coating failure? Reference: Corrosion and Coating, The Society for Protective Coatings Answers on page 12 J A N U A RY 2007 I N T E R FA C E • 1 1 ing the relative performance of EPDM, thermoplastic, modified bitumen, and built-up roofing systems; the relative similarity of these systems in terms of EUAC may indicate that no single system offers an unassailable economic advantage. Perhaps this is why each of these major approaches to roofing enjoys a respectable share of today’s commercial roofing market. EDITOR’S NOTE: A copy of the Excel workbook used to calculate life cycle cost using the Equivalent Uniform Annual Cost (EUAC) method may be obtained by e-mailing Jim Hoff at hoffjames@firestonebp.com. This paper was originally presented at the RCI 21st International Convention and Trade Show in Phoenix, Arizona, in March 2006. REFERENCES Boon, R. A., “Better Roofs are Less Expensive,” Roofing Contractor, November 2001. Cash, C.G., “The Relative Durability of Low-slope Roofing,” Proceedings of the Fourth International Symposium on Roofing Technology, pgs. 119- 124. National Roofing Contractors Association: Rosemont, IL, 1997. Fuller, S. K., and Rushing, A. S., “Energy Price Indices and Discount Factors for Life-cycle Cost Analysis,” April 1, 2005 to March 31, 2006, Annual Supplement to NIST Handbook 135 and NBS Special Publication 709 (NISTIR 85-3273-20). Gaithersburg, MD: National Institute of Standards and Technology, 2005. 2004-05 Low-slope Roofing Materials Guide, Rosemont, IL: National Roofing Contractors Association, 2005. Means Building Construction Cost Data, Kingston, MA: R.S. Means, 2005. Schneider, K.G., and Keenan, A.S, “A Documented Historical Performance of Roofing Assemblies in the United States, 1975-1996, Proceedings of the Fourth International Symposium on Roofing Technology, pgs. 132- 137, Rosemont, IL: National Roofing Contractors Association, 1997. “White Paper on Sustainability,” Building Design & Construction, November supplement, 2003. Chart 1: Equivalent Uniform Annual Cost (EUAC) of Roofing Systems Answers to questions from page 11: 1. • Synergistic effect • Aesthetics • Added protection • Color coding • Safety markings 2. Aesthetics 3. • Pigment – provides body. • Binder – provides important film properties. • Solvent – reduces viscosity for easy application. 4. • Barrier protection. • Inhibitive pigment protection. • Sacrificial protection. 5. • Water traps – configurations with pockets that collect water. • Sharp edges – coatings retract from sharp edges, leaving only film. • Crevices – at bolted seams and back-to-back angle iron. • Dissimilar metals – accelerated corrosion of more chemically reactive metal. • Areas difficult to access – lack of sufficient coating. Jim Hoff has served in a variety of technical and management roles in the construction industry for over 30 years. Currently, Mr. Hoff is vice president of technology and product development for Firestone Building Products Company. He received an A.A.S. from Indiana Vocational Technical College, a B.A. in psychology from Indiana University, a M.S. in management from Indiana Wesleyan University, and currently is completing his doctoral dissertation for a D.B.A. in management from the University of Sarasota. Mr. Hoff serves as a board member of the RCI Foundation. James Hoff 12 • I N T E R FA C E J A N U A RY 2007
