The Issue Should the Board of Professional Engineers impose discipline against Nicholas Nicholson, P.E., for engaging in negligence in the practice of engineering, in violation of Section 471.033(1)(g), Florida Statutes?
Findings Of Fact Respondent is and has been at all times material to this matter, a licensed professional engineer in the State of Florida, having been issued License No. PE 34603. Respondent was the engineer of record for plans for a Sweetwater Homes of Citrus model home located in Citrus County, Florida. On November 12, 1997, Respondent signed and sealed drawings for Sheet 1 of 4 Floor Plan (Model Tradewinds R), Sheet 2 of 4 Foundation and Plumbing Plan (Model Tradewinds R), and Sheet 4 of 4 Elevations (Model Tradewinds Lc) for the Sweetwater Homes of Citrus. On November 17, 1997, Respondent signed and sealed drawings for Sheet 1 of 1 Typical Wall Section and Details. On December 18, 1997, Respondent signed and sealed drawings for an untitled truss layout sheet. Respondent certified all five sheets with the statement: "I hereby certify that I have performed the attached design to comply with 100 mph loads and it is in compliance with Section 1606 of the Standard Building Code." The sheets were submitted to the Citrus County Building Department for permitting. In Sheet 4 of 4, the elevation, as shown, is inconsistent with the indicated roof framing in that, contrary to what is shown on the Elevation, beginning at a point 30'-6" from the rear of the building, there is a 14'-2" length of exterior wall which is 15'-0" high. The bearing wall between the Garage and Bedroom No. 3 is not a shear wall. On top of the bond beam is a gable truss which at the ridge rises some 10'-5" higher. (Respondent does not dispute this fact but prefaces the admission to the extent that the 10'-5" reference is an approximation.) In the 44'-4" clear distance between shear walls, there are seven #5 vertical bars in grouted cells. ADDITIONAL FACTS The residence in question is the Morse residence. A Certificate of Occupancy for the residence was issued as required by the Codes and Ordinances of Citrus County in relation to required inspections performed by county building officials. The date of issuance was July 10, 1998. The drawings signed and sealed by Respondent for the Morse residence that formed the basis for the Administrative Complaint were admitted as Petitioner's Exhibit No. 4. The items found within the plans that Petitioner claims constitute negligence in the practice of engineering, disputed by Respondent, are divided into two categories. The first category concerns what is referred to in the Administrative Complaint as the "gable end-left side" as depicted in the elevation in Sheet 4 of 4, in particular the left elevation. The complaint in this category concerning the gable end-left side elevation is further specified in relation to the consistency of the depiction in Sheet 4 of 4 of the elevation when contrasted with the depiction of the roof framing plan on Sheet 4 of 4, in that beyond the 30'-6" point from the rear of the building, there is a 14'-2" length of the left elevation exterior wall which is 15'-0" high. The scale drawing of the left elevation does not describe the 14'-2" section of the exterior wall to be 15'-0" high, as contemplated by the roof framing plan. While Respondent concedes the inconsistency between the indicated roof framing and left elevation on Sheet 4 of 4, he contests whether this deficiency causes him to be out of compliance with the expected standards for engineering practice. James Owen Power, P.E., is a registered engineer in Florida. He holds a bachelor of science degree in civil engineering. He was recognized as an expert in structural engineering in this proceeding. Mr. Power confirmed the height of the wall in the 14'-2" section by examining other elevations depicted aside from the left elevation. He resorted to the truss layout sheet for the residence to support his understanding. He expressed the concern that the discrepancy between the depiction of the wall in the left elevation and the true height of the wall would have a consequence in the ability of the wall to resist wind loads. Mr. Power identified the wall height where the 14'-2" section is found as promoting the possible failure to provide adequate bracing for the top of the block wall and the prospect that the tie beam atop the wall would not continue from the 30'-6" section through the 14'-2" section of the wall. Wilbur T. Yaxley, P.E., graduated from a five-year program for structural engineering. He is a registered professional engineer in Florida. He was recognized as an expert in structural and forensic engineering. He does not take issue with the fact that the left-side elevation on Sheet 4 of 4 fails to indicate the proper height of the 14'-2" section of the exterior wall. He considers the error to be a simple mistake that does not meaningfully impugn Respondent's performance as an engineer. Petitioner singled out the discrepancy between the depiction of the roof framing and the left elevation as drawn in relation to the 14'-2" section of the exterior wall, as constituting an actionable circumstance. As such, this mistake has not been shown to have meaningful significance in the disciplinary context. The balance of the allegations under the heading, "gable end-left side (elevation Sheet 4 of 4)", pertain to the subject of wind loads on the left elevation exterior wall and the adequacy of the wall to resist wind loads, as portrayed in the plans, consistent with sound engineering practices, recognizing that Respondent has certified that the design complied with 100 mph loads described in Section 1606 of the Standard Building Code. At the time in question the Standard Building Code in effect was the 1994 version. Within that document is Section 1606, entitled "Wind loads." Based upon the reading of that provision and other reference sources deemed to have pertinence, Petitioner, through its consultant experts, expresses the belief that the design is unable to withstand the 100 mph wind load. Further, Petitioner asserts in its proof that to properly respond to wind loading, the plans must not only address the main wind force resisting system requirement but that of components and cladding and the ability of those elements to resist wind loads when designed for separate load cases. Petitioner's position is supported by the opinion testimony of Mr. Powers. It also relies on the opinion testimony of Peter J. Iglesias, P.E., who is licensed in Florida and was recognized as an expert in structural engineering and plans examination. In opposition to Petitioner's position, Respondent and his expert, Mr. Yaxley, contend that the use of a computer program by EDA Software, Inc., to calculate main wind force resisting coefficients solely, as a means to respond to wind loads, is recognized in Florida and that the software was properly exercised for that purpose in designing the Morse residence. The respective opinions by the parties in their legitimacy are discussed in the interest of resolving the dispute. Petitioner's experts read Section 1606.2.2.2., Standard Building Code 1994, as establishing the necessity in design to calculate not only wind loads related to the main wind force resisting system, but wind loads for components and cladding in the instance where those members act as part of the main wind force resisting system and as individual components in cladding. In those cases, separate loads shall be calculated. Section 1606.2.2.2 states, "members that act as both part of the main wind force resisting system and as components in cladding shall be designed for separate load cases." References are also made to the definitions within Section 1606.2.3, Special Definitions for 1606.2.: For the purpose of this section, certain special terms are defined as follows: COMPONENTS AND CLADDING-elements that are either directly loaded by the wind or receive wind loads originating at relatively close locations and that transfer these loads to the main wind force resisting system. * * * MAIN WIND FORCE RESISTING SYSTEM (MWFRS) - An assemblage of major structural elements assigned to provide support for secondary members in cladding. The system primarily receives wind loading from more than one surface. Both parties recognize that the calculations in association with a main wind force resisting system predicts lesser wind loads than the use of components and cladding coefficients to determine wind loads, thus creating different expectations in the design to meet expected wind loads. Respondent in the present design did not exercise the coefficients for components and cladding when addressing wind loads. To counter Petitioner's contention Respondent asserts, through his proof, that there is no consensus among engineers in various fields concerning the need to use components and cladding coefficients or limit the consideration to main wind force coefficients when determining wind loads for design purposes. Respondent defers to correspondence dated August 23, 1995, from Larry M. Schneider, Chairman of the Florida Board of Building Codes and Standards, to a Mr. Edward Hubert, P.E., which Respondent believes justifies the use of the EDA Software program to determine wind loads in the manner Respondent employed. That correspondence states in pertinent part: The Florida Board of Building Codes and Standards, as authorized by Section 489.115(4)(b)2., Florida Statutes, recognizes the EDA Software, Inc., Residential Wind Design Analysis computer program as an alternative method for demonstrating compliance with Section 1606, Wind Loads of the Standard Building Code. The Board recognizes this methodology for designs of one and two family dwellings developed by this method when conducted by Florida Registered Engineers as authorized by Section 471.015, Florida Statutes, Registered Architects as authorized by 481.213, Florida Statutes, and Division I contractors as authorized by Section 489.115 or Section 489.117, Florida Statutes, and when submittals for permitting include sufficient details and specifications (such as that provided on the computer program summary printout) for evaluation and inspection by the permitting jurisdiction. The EDA Software for Residential Wind Design Analysis employed by Respondent in determining the main wind force resisting system results for 100 mph wind loads is recognized in his field. There was no showing that he failed to properly exercise the software. He made separate calculations for the rat runs in the design as they influenced the main wind force resisting system calculations for the residence. Respondent also relies on a declaratory statement in the matter of Edward F. Hubert, P.E., Petitioner, DCA 99-DEC- 046, before the State of Florida Building Commission, in its remarks concerning the EDA Software for wind design as a means to justify the choice to employ the software in designing the Morse residence; however, the declaratory statement was issued on April 20, 1999, beyond the point at which the drawings were rendered. Thus, the declaratory statement created could not justify the choice to use the computer software. Finally, Respondent, in its proof, urges that the Petitioner's criticism of the design tends to ignore certain elements within the design of the structure which would lend strength to this residence in resisting wind loads. With this general background discussion in mind, the remaining allegations are examined. Petitioner alleges in paragraph 10b that: The rear of the building 10'-0" above the finished floor is an 8" x 8" grouted bond beam reinforced with one No. 5 reinforcing bar. There is no intermediate structural element capable of providing adequate lateral resistance; the bearing wall between the Garage and Bedroom No. 3 is not a shear wall. The plan does not make clear whether the wall height is 10'-0" above the finished floor. If the wall height is 8'-0", excluding window or door openings, an 8" x 8" tie beam with one #5 reinforcing bar would be envisioned. If the wall is 10'-0", then a 16'-0" tie beam would be used with two #5 reinforcing bars, according to the plan. The gravamen of the complaint concerning this allegation is in the instance where there is available an 8" x 8" tie beam (grouted bond beam) with one #5 reinforcing bar associated with the wall in question, lacking some intermediate structural element to provide lateral resistance to wind loading, excluding the bearing wall between the garage and bedroom No. 3, which is not a shear wall, the design fails to meet the standards for engineering practice. Mr. Power expresses the opinion that the lateral resistance that is present with the design elements either in their individual capacity or working together do not suffice to meet the 100 mph wind load. While Mr. Yaxley agrees that the wall between the garage and bedroom No. 3 is not a shear wall allowed to support the outside masonry wall, he is persuaded that the placement and spacing of the "rat run" associated with the outside wall, on the top of the bottom cord of the trusses, would create lateral resistance. Mr. Yaxley also refers to the filled cells with rods that extend through the block wall from where they are buried in the foundation up to the point where they are tied into the tie beam, creating a column to resist wind loading. In arriving at his opinion concerning the support for the wall in resisting lateral wind loads, Mr. Yaxley has referred to source material from the Standard Building Code Congress International. Mr. Power was mindful of the contribution that the cells filled with concrete would make to providing lateral support for the outside wall in question when expressing the opinion that the available structural elements were not adequate to resist lateral wind loading. The number of grouted cells discussed by Mr. Yaxley and Mr. Power in the subject wall is seven. Petitioner alleges at paragraph 10c that: On top of the bond beam is a gable truss which at the ridge rises some 10'-5" higher. The bond beam, acting laterally cannot safely span the 44'-0" clear distance between the shear walls when subjected to code specified wind loads from wall and gable truss. Mr. Power expresses the opinion that the gable end truss has no adequate capacity to brace the top of the block wall. It provides no lateral support, in his view. The connection between the gable truss and the masonry wall is a hinge at the top of the block wall and the bottom of the gable end truss. Lacking the ability of the gable end truss to adequately brace the top of the block wall which spans the 44'-4" between the opposite end shear walls, some other source of support would be necessary to resist wind loading. Mr. Power suggests the possibility of the use of a ceiling diaphragm for that purpose, but does not believe the drawings make reference to a ceiling diaphragm. In addition to the need to state, call out, or detail reference to the use of a ceiling diaphragm in the plan, Mr. Power expressed the opinion that the Standard Building Code would not allow a ceiling diaphragm of gypsum board. Gypsum Board was the anticipated ceiling material in the plan. According to Mr. Power, the imposed wind load on the wall is 24 1/2 pounds per square foot, but the wall acting as a cantilever can only resist wind loading at 6 1/2 pounds per square foot. Mr. Iglesias offers the opinion that the drawings do not specify the use of a ceiling diaphragm to allow for load transfer. Mr. Iglesias expressed the opinion that the multiple levels of ceiling within the residence defeated the purpose of the diaphragm in terms of the need for continuity, assuming that the ceiling to be installed could be considered a diaphragm. He, too, believes that there is a problem with the ability of the bond beam to safely span the 44'-4" clearance between shear walls when subjected to specified wind loads on the wall and gable truss. In his opinion the wall acts as a cantilever in the configuration where it lacks support at the top. Mr. Yaxley expresses the belief that the actual height of the ridge exceeds 10'-5", which would promote greater loading. He does not believe that the bond beam spans the entire 44'-4" distance without meaningful support. He perceives the distance to be roughly 30'-0", with the last 14'-0" subject to support provided by perpendicular trusses in the plan, as they are tied into the wall and extend back into the house offering lateral stability to the wall. The section of 14'-0" is an area where the trusses are perpendicular, with a truss anchor available for each truss. With this in mind, Mr. Yaxley believes there is sufficient support for the top of that wall in the 14'-0" section. Petitioner alleges at paragraph 10d that: In the 44'-4" clear distance between shear walls there are seven #5 vertical bars in grouted cells. The wall is incapable of safely spanning 10'-0" as a cantilever without lateral support at the level of the bond beam. Mr. Power perceives the seven #5 vertical bars in grouted cells as having value when the wall is cracked, in which case they would offer resisting capacity to the wall as a cantilever, but not as a member spanning from the foundation to the top of the wall, absent existing support at the top of the wall. Consequently, Mr. Power believes that the wall is incapable of safely spanning 10'-0" as a cantilever without that lateral support at the level of the bond beam. Taking into account the vertical reinforcing bars in the wall, the value of the bond beam or tie beam at the top of the masonry wall and 2 x 4 bracing in the truss base, Mr. Power does not believe that the system could accommodate 24.45 pounds per square foot of wind loads imposed per his calculations. Mr. Power believes that the safe capacity of the wall is 6.58 pounds per square foot, less than adequate. Mr. Yaxley differs in his opinion. He acknowledges the existence of the seven #5 vertical bars in grouted cells in the wall. If those filled cells constituted the only lateral support, Mr. Yaxley expressed the opinion that those vertical bars in grouted cells would not meet the 100 mph wind load requirement in the span. However, even absent specification of elements of a ceiling diaphragm in the plan, Mr. Yaxley believes that it is still acceptable to assume that a ceiling diaphragm would provide support for the wall. Mr. Yaxley expressed the opinion that under the Standard Building Code, there is a prohibition against the use of vertical diaphragms made of gypsum board. By inference, absent some reference within the Standard Building Code, prohibiting the use of gypsum board in horizontal plane as a diaphragm, he concludes that it can be used as a horizontal diaphragm. In addition, there are the aforementioned perpendicular roof trusses in the 14'-0" section of the wall which provides support. Mr. Yaxley made calculations concerning his belief that the wall was adequate to span the 10'-0" as a cantilever based upon the roof plywood diaphragm, the roof frame bracing, the tie beam and the wall itself. The second major category concerning the complaint is entitled, "Lateral Bracing - (Typical Wall Section and Details Sheet 1 of 1)." Within the second major category are found the following allegations: 10 e. The Code specifies an outward wind suction of 950 lbs., which load must be transferred in tension through the "rat run" and resisted by the diagonal 2 x 4 strut, resulting in compressive force of 1,467 lbs. in that member which is unbraced over its 12'-4" length and is thus grossly overstressed. 10 f. The detail does not specify the essential connections and there is no feasible way to provide for an adequate stress transfer in the arrangement shown. 10 g. To transfer 1,467 lbs. safely, at least nine nails would be required. They are not specified in the plans and they cannot effectively be installed in the arrangement as shown. Mr. Power expresses the opinion that the system design, in which diagonal struts are involved in the transfer of loads up to the roof sheeting under conditions of suction, will not work. According to calculations performed by Mr. Power the compressive force exerted on those struts can be as little as 1,369 pounds or as much as 1,467 pounds. The wind load on the tributary area involved in this analysis can be as little as 920 pounds or as much as 950 pounds inward or outward at the rat run, per the calculations, at the interface of the block wall and the gable end truss. Although the rat run is nailed to the trusses, the trusses do not support the rat run. If 920 pounds is exerted, acting outward or under suction, it is delivered to the bottom of the diagonal member, which Mr. Power presumes to have been connected at either end of the diagonal member. With this transfer a condition would be created which grossly overstresses the diagonal member. Mr. Power does not believe that the diagonal member would act effectively under those circumstances. Although no details are offered concerning the connections of the diagonal member in the plans, Mr. Power calculates that as few as eight or as many as nine nails would be needed to provide for adequate stress transfer. Based upon the size of the member, the diagonal brace, eight or nine nails could not fit into the space available for the connection to accommodate transfer resulting from compressive force. Mr. Yaxley questions Mr. Power's calculation of the amount of wind suction created as a function of the use of the 1994 Standard Building Code. By his calculations Mr. Yaxley believes that the suction is 126 pounds. Mr. Yaxley does not believe that the diagonal member in the plan is grossly overstressed. Mr. Yaxley does not believe that the diagonal member is necessary to resist wind loads, all other features in the wind force resisting system being known. By his calculations, Mr. Yaxley has concluded that there is sufficient stiffness in the masonry wall, the ceiling, and the rat runs to support the wall without the use of the diagonal struts. Mr. Yaxley acknowledges that the nature of the attachment of the diagonal strut is not detailed in the plans but believes that the plans contemplate some form of attachment to the bottom chord of the truss. Lacking specificity as to how many nails would be used to make the attachment, Mr. Yaxley relies on his general knowledge that two or three nails would be employed. Hypothetically, assuming a load transfer in the amount of 1,467 pounds, Mr. Yaxley believes that seven nails would be sufficient to connect the diagonal strut and resist the compressive force. On a related topic, Mr. Yaxley's inquiries and his reading of the specifications associated with the TA 18 truss anchors called for in the plan lead him to believe those truss anchors provide three-directional resistance to forces. Those features of resistance which are available individually concern up-lift, resistance lateral to the wall and resistance parallel to the wall, without diminished effect when employed in the three capacities. Mr. Iglesias expresses the opinion that the TA 18 truss anchors, based upon his discussions with others and his understanding of the TA 18 truss anchors uses and their ability to resist loads in more than one direction, could not perform in that manner. Mr. Power expresses the ultimate opinion that Respondent did not exercise due care in performing the function of an engineer. The underlying basis for the opinion concerns the failure to provide adequate resistance to brace the wall in question, having certified that the wall was in conformance with the 1994 Standard Building Code for 100 mph wind loads. Likewise, Mr. Iglesias finds fault with Respondent's performance. Mr. Iglesias does not believe Respondent utilized due care in performing in an engineering capacity in that certification was made that the plans complied with the 100 mph wind standard under the Standard Building Code, and Mr. Iglesias does not believe that to be correct. In opposition, Mr. Yaxley believes that Respondent's plans did satisfy the requirements of the Standard Building Code in relation to the 100 mph wind loads. Mr. Power, in his underlining opinions, posited the view that wind loading is not imposed evenly over building surfaces, such as the wall in question. He believes there are major fluctuations in intensity from one point to the next with occasional loads of extreme intensity concentrated over relatively small areas. This means that while trying to examine the effects of wind loads on the entire building or some major portion, such as the wall in question, the use of average wind loads is an acceptable approach but does not address the more intense concentration of wind loads on small areas, which cannot be ignored. These differences are addressed in the Standard Building Code, envisioned by Section 1606.2.2.2, under measures for the main wind force resisting system in the entire building or a major portion and for components and cladding for smaller areas of concentration as separate load cases. Under Section 1606.2.3, in its definition of components and cladding, Mr. Power believes that the exterior wall in question is an element directly loaded by the wind, receiving wind loads originating in relatively close locations and transferring those loads to the main wind force resisting system, calling for a consideration of the wind forces consistent with Section 1606.2.5 and its more stringent requirements for coefficients, as well as the less stringent coefficients associated with Section 1606.2.4 related to main wind force resisting systems. Mr. Iglesias had a similar opinion to Mr. Power's in expressing the belief that components and cladding needed to be examined as part of the process in evaluating the adequacy of the plans. In this instance, Mr. Iglesias believes that components and cladding coefficients should be applied to examine smaller areas in the wall. Mr. Iglesias expresses the opinion that certain structural elements may be considered part of the main wind force resisting system and also a component and cladding, requiring the calculation of coefficients under both approaches in the wall in question, a structural element involved in both calculations of the effects of wind loading. To try to gain an understanding concerning the question of whether a consideration of wind loads should be limited to an evaluation of the main wind force resisting system, Mr. Yaxley spoke to persons within the profession who he felt might provide insight and ultimately determined that in his opinion, the Standard Building Code, when examining questions of wind loads for the subject wall, would be sufficiently addressed by calculations for the main wind force resisting systems pursuant to those coefficients. For Mr. Yaxley, to determine the adequacy to resist the 100 mph wind load is accomplished by examining coefficients for the main wind force resisting system on average and not small areas under the coefficients associated with components and cladding. Factual Determination The August 23, 1995, correspondence from the Chairman of the Florida Board of Building Codes and Standards authorized the use of the EDA Residential Wind Design Analysis computer program. The letter of authorization did not identify how that software was to be employed when examining the requirements of Section 1606 of the Standard Building Code. Respondent was entitled to use the EDA Software to determine wind loads but he must exercise that tool consistent with the proper interpretation of Section 1606 as written. To that end, there are elements within the subject wall which act both as parts of the main wind force resisting system and as components and cladding. This made it necessary for the design to contemplate the more rigorous influence of wind loads under coefficients associated with components and cladding. It was not enough to limit consideration to coefficients involved with wind loading under the main wind force resisting system. The Residential Wind Design and Analysis computer program lends itself to both an assessment of coefficients associated with the main wind force resisting system and those coefficients pertaining to components and cladding. Alternatively, without intending to consider wind loading on effected components and cladding as a function of the computer software, there was the prospect that the plan would meet those requirements for wind loading coincidentally. This refers to the elements available in the subject wall over its expanse for lateral resistance measured against wind loads associated with components and cladding. As detailed in the plan, the tie beam, the #5 vertical bars in grouted cells, the wall if devoid of reinforcement, the trusses with anchors, and the rat run with its diagonal support were not sufficient to meet the components and cladding wind loads. Moreover, TA 18 truss anchors cannot serve the several purposes for accommodating loads at once, as Respondent contends in his proof. The nature of the connection of the diagonal braces to other structural elements is not stated in the plans. The nature of the lateral bracing is not conducive to being connected in an effective manner to accommodate stress transfer. A ceiling diaphragm is not detailed. The ceiling material could not serve as a diaphragm. The ceiling integrity would not be effective as a diaphragm. For these reasons, Respondent's performance was less than that expected of an engineer practicing in Florida.
Recommendation Based upon the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED: That Respondent receive a written reprimand and pay a fine in the amount of $1,000 and costs be assessed in accordance with Section 455.227(3)(a), Florida Statutes. DONE AND ENTERED this 28th day of January, 2002, in Tallahassee, Leon County, Florida. CHARLES C. ADAMS Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 SUNCOM 278-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us Filed with the Clerk of the Division of Administrative Hearings this 28th day of January, 2002. COPIES FURNISHED: David P. Rankin, Esquire The Law Offices of David P. Rankin, P.A. 3837 Northdale Boulevard, Suite 332 Tampa, Florida 33624 Douglas Sunshine, Esquire Florida Engineers Management Corporation 2507 Callaway Road, Suite 200 Tallahassee, Florida 32303 Natalie A. Lowe, Executive Director Board of Professional Engineers Florida Engineers Management Corporation 2507 Callaway Road Tallahassee, Florida 32303-5267 Hardy L. Roberts, III, General Counsel Department of Business and Professional Regulation 1940 North Monroe Street Tallahassee, Florida 32399-2202
The Issue The issue is whether Petitioner is entitled to reimbursement of its attorneys' fees and costs under section 120.595(3), Florida Statutes, from its successful prosecution of a rule challenge in DOAH Case No. 12-1944RX.
Findings Of Fact In DOAH Case No. 12-1944RX, Petitioner challenged Florida Administrative Code Rules 69O-170.017 and 69O-170.0155 and incorporated forms OIR-B1-1699 (Form 1699) and OIR-B1-1655 (Form 1655). These rules generally relate to fixtures and construction techniques that mitigate wind loss and earn homeowners a discount on the wind portion of their homeowners' insurance premium. The most important of these rules, Form 1699 consists of two matrices: one matrix provides discounts for mitigative fixtures and construction techniques applied to existing residential construction (i.e., predating the 2001 Florida Building Code), and the other matrix provides discounts for mitigative fixtures and construction techniques applied to new residential construction (i.e., subject to the 2001 Florida Building Code or any of its successors). The Final Order invalidates the matrix applicable to existing residential construction, but not the matrix applicable to new residential construction. The Final Order concludes that the omission from Form 1699 of discounts for increased wind resistivity for doors modifies and contravenes the law implemented and is arbitrary. The Final Order finds other omissions from Form 1699--i.e., discounts for increased wind resistivities for windows and increased impact resistivities for doors--but these findings served the purpose of partly justifying the invalidation of the entire form for the omission of a single set of discounts substantially affecting Petitioner--i.e., the discounts for increased wind resistivities for doors. (Findings as to the interdependency of all of the discounts provided the remaining justification for invalidating the entire form for the omission of a single set of discounts.) Form 1655 advises homeowners of the availability of discounts applicable to the wind portion of their homeowners' insurance premiums for various mitigative fixtures and construction techniques. The Final Order invalidates Form 1655 in its entirety. The main reason is that Form 1655 fails to notify homeowners about the availability of discounts for fixtures and construction techniques that increase the wind resistivity of windows and doors. An additional reason is that Form 1655 mentions shutters as the sole fixture to increase the impact resistivity of windows and doors, misleadingly implying that shutters are the sole fixture or construction technique for increasing the impact resistivity of windows and doors. Rule 69O-170.017 incorporates by reference Form 1699. The Final Order denied Petitioner's request to invalidate rule 69O-170.017 because the rule incorporates the still-valid, existing-construction matrix in Form 1699. Rule 69O-170.0155 incorporates by reference several forms. The Final Order invalidates only rule 69O-170.0155(k), which is the subsection that incorporates Form 1655. Petitioner commenced its rule challenge to obtain a wind-premium discount for homeowners who purchased and installed its bracing system on their existing, nonglazed garage doors in order to increase their wind resistivity. The thrust of Petitioner's challenge was thus to the omission from Form 1699 of any discounts for fixtures and construction techniques that increase the wind resistivity of doors. Obviously, the shortcomings of Form 1655--and its adopting rule, rule 69O-170.0155(k)--were almost entirely derived from this omission from Form 1699. Respondents' liability for attorneys' fees and costs thus requires consideration only of its adoption of Form 1699 without any discounts for fixtures and construction techniques that increase the wind resistivity of doors. As noted in the Final Order, the establishment of discounts for all mitigative fixtures and construction techniques is a complicated process. The actuarial expertise necessary to complete this task resides in Respondent Office of Insurance Regulation (OIR), but is itself dependent on engineering expertise that is not found within either respondent, or at least was not in 2006 when Form 1699 was adopted. The engineering work underlying Form 1699 featured computer modeling, among other things, to project the salient features of storms that may be expected to strike various parts of Florida over thousands of years; as for impact resistivity, to project the trajectories and momentum of missiles that will be launched by these storms; to place in the path of these storms and missiles various forms of residential construction with relevant combinations of mitigative fixtures and construction techniques covering several factors, including the protection of windows and doors from impacts and the protection of windows and doors from wind (without regard to impacts); to project the damage states that will result from these modeled storms upon individual hypothesized residential buildings; and to project the economic losses--with particular emphasis on insured losses--that will result from these damages. The relevant timeframe for this case begins with Hurricane Andrew in 1992. As the Final Order describes, the Florida legislature and other federal and state agencies and organizations reacted swiftly and comprehensively to this storm and the catastrophic damage and loss that it caused. FEMA quickly published its analysis of, among other things, the relationship between construction and storm damage. In 2001, the legislature adopted the Florida Building Code (FBC), which required, among other things, new construction to meet wind loads specified in the code, based on projected wind speeds in different regions of Florida. Almost at the same time that the 2001 FBC went into effect, in March 2002, Applied Research Associates, Inc., published the Development of Loss Relativities for Wind Resistive Features of Residential Structures (2002 ARA Report). Procured by the Florida Department of Community Affairs, which, at the time, had considerable responsibilities in the adoption of the 2001 FBC, the 2002 ARA Report was a groundbreaking achievement in modeling the effects, in terms of reduced damage and loss, from various forms of mitigative fixtures and construction techniques, alone and in almost countless combinations. For present purposes, the focal point of the 2002 ARA Report were tables of loss relativities, which provided factors by which to calculate how different combinations of mitigative fixtures and construction techniques reduced wind losses. Taking these data, OIR's actuaries issued in January 2003 an informational memorandum and a precursor to Form 1699, which suggested premium discounts to be used by homeowners' insurers when filing insurance rates. (Then and now, insurers are permitted to use other data sources in setting their rates, but all but two of them use the suggested discounts in Form 1699.) In August 2004, Hurricane Charley struck Florida. A design wind event, like Hurricane Andrew, the timing of Charley, after the adoption of the 2001 FBC, proved the effectiveness of the 2001 FBC in requiring fixtures and construction techniques that demonstrably mitigated wind damage and loss. In 2006, respondents issued Form 1699 in its present form, eliminating a dampening factor that they had included in the precursor form three years earlier. (To allow insurers to adapt to the new rate-setting environment, respondents had halved the discounts in the precursor form.) The 2002 ARA Report claims to adhere to the statutory mandate contained in section 627.0629(1), to determine discounts for fixtures or construction techniques that "enhance roof strength, roof covering performance, roof-to-wall strength, wall-to-floor-to-foundation strength, opening protection, and window, door, and skylight strength." For reasons explained in the Final Order, "opening protection" is limited to the impact resistivity of windows and doors, and the "strength" of windows and doors (skylights being treated as windows) is limited to their wind resistivity. Unfortunately, the 2002 ARA Report collapsed opening protection and the strength of windows and doors into one category--opening protection--so ARA never developed loss relativities for fixtures and construction techniques that increased the wind resistivities of doors or, for that matter, windows. As noted above, respondents were entirely dependent on the work of ARA due to its specialized knowledge of the FBC and, more generally, its expertise in engineering and computer modeling. ARA, not respondents, possessed this highly specialized knowledge, which was necessary to generate the loss relativities, on which respondents, in turn, could rely to generate the legislatively mandated premium discounts. The omission of loss relativities for the strength of windows and doors--as a standalone category or within the category of opening protection--is not apparent in the richly detailed 2002 ARA Report. The above-described facts--coupled with the time- pressured nature of the task assigned to respondents--provide the reasonable basis in fact for the adoption of the portion of Form 1699 that has been invalidated. The factual justification for the adoption of the portion of Form 1699 that has been invalidated is greater than any legal justification that respondents may claim. The statute truly is a model of clarity--and succinctness. Reduced to its plainest terms, the statute calls for discounts for six categories of mitigative fixtures and construction techniques, and ARA and respondents addressed only five. However, some legal justification exists for the adoption of the portion of Form 1699 that has been invalidated. First, the legal mandate of section 627.0629(1) does not exist in a vacuum; it operates in the complex facts of engineers, computer programmers, and actuaries whose work is necessary to lend meaning to the statutory mandate. To this extent, respondents find some legal justification for the same reason that they find ample factual justification for the adoption of the portion of Form 1699 that has been invalidated. Second, the legislature itself missed a clear, early opportunity to remind respondents of their failure--obvious, perhaps, only in hindsight--to address the omitted sixth factor enumerated in section 627.0629(1). The precursor of Form 1699 likewise omitted discounts for fixtures and construction techniques that increased the wind resistivities of doors and windows. When, in 2006, the legislature mandated the adoption of full discounts, without any dampening, it easily could have forcibly reminded respondents that they--and their contractor-- had missed one of the six statutory discounts. The subtlety of respondents' legal error seems to have eluded the legislature, as well. Third, even in hindsight, the legal underpinning of the invalidation of the existing-construction matrix of Form 1699 is sometimes elusive, given the temptation to join ARA and respondents in analyzing wind resistivities under the factor of opening protection. As disclosed at the hearing, the Administrative Law Judge spent a considerable amount of time, in preparing the Final Order, misanalyzing respondents' treatment of the wind resistivities of doors from the perspective of opening protection. Repeated, close readings of section 627.0629(1), in the context of the complex materials presented in the 2002 ARA Report, eventually revealed the now-clear legal principle that the omitted sixth statutory factor--the strength of windows and doors--applied to wind resistivity (and opening protection was restricted to impact resistivity). And Petitioner itself joined in exactly the same misanalysis, both in its pleading and proof at the hearing in the rule challenge. Seeming to yield once more to this misanalysis, even in the fee hearing, Petitioner cross-examined OIR's lone witness with an emphasis on respondents' flawed decision, as described in the Final Order, to omit a discount for doors under opening protection.
The Issue The issues are whether Florida Administrative Code Rules 69O-170.017 and 69O-170.0155 and incorporated forms OIR-B1-1699 (Form 1699) and OIR-B1-1655 (Form 1655) are invalid exercises of delegated legislative authority, pursuant to section 120.56(3), Florida Statutes, on the grounds that they enlarge, modify, or contravene the specific provisions of the law implemented or that they are arbitrary or capricious, as prohibited by section 120.52(8)(c) and (e), respectively.
Findings Of Fact Rules, Forms, and Relevant Law Implemented Citing section 627.0629, Florida Statutes, among other statutes, as the law implemented, Florida Administrative Code Rule 69O-170.0155(k) incorporates by reference Form 1655. Form 1655 advises homeowners of the availability of discounts applicable to the wind portion of their homeowners' insurance premium for various fixtures and construction techniques that mitigate wind loss. As relevant to this case, Form 1655 notifies a homeowner of a discount for "shutters." Although this discount implicitly applies to windows and possibly doors, its levels of protection--"none," "intermediate," and "hurricane"--indicate, for the reasons discussed below, that the discount is available only for a fixture--i.e., shutters--that increases the impact resistivity, not wind resistivity, of windows and doors.1/ Impact resistivity and wind resistivity are defined in the first paragraph of the next part of this Final Order. Form 1655 notifies a homeowner that she may qualify for multiple discounts, if her home is mitigated with multiple fixtures or construction techniques listed on the form. However, Form 1655 warns that discounts are not calculated cumulatively: "[w]hen one discount is applied, other discounts are reduced until you reach your maximum discount." Citing section 627.0629(1), among other statutes, as the law implemented, Florida Administrative Code Rule 69O-170.017 incorporates by reference Form 1699. Rule 69O-170.017 requires insurers to base their rates on the discounts contained in Form 1699 "without any modification unless [the insurer's rates] are supported by detail[ed] alternate studies where all assumptions are available to [OIR] for review." Form 1699 consists of two pages. On the first page is a matrix of wind-premium discounts for various combinations of mitigative fixtures and construction techniques applicable to "existing construction."2/ All references in this Final Order to Form 1699 are to the first page of the form. Containing 12 columns3/ and 72 rows,4/ Form 1699 identifies discounts under eight columns5/ and along 72 rows, for a total of 576 discounts.6/ The column labeled "opening protection" has three levels: "none," "Basic--Windows or All," and "Hurricane--Windows or All." Form 1699 lacks a column or row for the strength of windows, doors, and skylights. Discounts for fixtures and construction techniques that increase opening protection and window, door, and skylight strength are required by section 627.0629(1): It is the intent of the Legislature that insurers provide savings to consumers who install or implement windstorm damage mitigation techniques, alterations, or solutions to their properties to prevent windstorm losses. A rate filing for residential property insurance must include actuarially reasonable discounts . . . for properties on which fixtures or construction techniques demonstrated to reduce the amount of loss in a windstorm have been installed or implemented. The fixtures or construction techniques must include, but are not limited to, fixtures or construction techniques that enhance roof strength, roof covering performance, roof-to-wall strength, wall-to-floor-to-foundation strength, opening protection, and window, door, and skylight strength (emphasis added). [D]iscounts for fixtures and construction techniques that meet the minimum requirements of the Florida Building Code must be included in the rate filing. [OIR] shall determine the discounts . . . that reflect the full actuarial value of such revaluation, which may be used by insurers in rate filings. As to the two rules and two forms at issue in this case, OIR initiated rulemaking and made recommendations to Respondent Financial Services Commission, which, as noted in the Conclusions of Law, is the agency authorized to adopt these rules. Basic Principles and Overview Hurricane winds damage structures by wind pressure and wind-borne debris. Wind-borne debris, which is also known as missiles, consists of such objects as roof tiles or tree branches that are transported by the wind into roofs, exterior walls, windows, and doors, sometimes penetrating these building components. The ability of a building or building component to resist wind pressure is a measure of its wind resistivity, and the ability of a building or building component to resist wind- borne debris is a measure of its impact resistivity. Both wind and wind-borne debris may damage the roof, exterior walls, windows, doors, and--rarely, in the case of hurricanes--foundations.7/ These five components form the building envelope. Once the building envelope is breached, rainwater may enter and damage the interior of the building; more dramatically, hurricane-force winds may pressurize the interior of the building, blowing out windows, doors, or exterior walls or blowing off the roof.8/ The identification of premium discounts is the culmination of a complicated process. Discounts are driven by projections of damage9/ and economic loss.10/ Damage and economic loss are driven by the interaction of wind pressure and wind-borne debris from simulated hurricanes with structures that are modeled with various wind resistive and impact resistive features. At every step of this overall process, difficult calculations are required, and some of the results may defy a layperson's intuitive expectations.11/ The data and analysis on which OIR relied in identifying the discounts in Form 1699 were derived from a study reported in Applied Research Associates, Inc. (ARA), Development of Loss Relativities for Wind Resistive Features of Residential Structures (Version 2.2 March 2002) (2002 ARA Report). Another important source of data is a followup study reported in ARA, Florida Residential Wind Loss Mitigation Study (Version 1.11 October 2008) (2008 ARA Report), although OIR has not revised the discounts in Form 1699 based on the data and analysis contained in the 2008 ARA Report. Openings in structures are framed-out sections of a wall or roof--most commonly for windows and doors, including garage doors. The protection of openings helps preserve the integrity of the building envelope for all structures relevant to this case.12/ Hurricane losses are governed by the performance of the building envelope,13/ especially once the winds exceed 120 mph.14/ All components--i.e., windows, skylights, and doors, including garage doors--installed in openings are glazed or nonglazed. A component is glazed if it contains any glass or plastic. All windows and skylights are glazed, as are doors with glass or plastic windows. A component is nonglazed if it contains no glass or plastic. This Final Order refers to glazed components in openings as windows and generally refers to nonglazed components in openings, including garage doors, as doors. All references to garage doors are to garage doors without windows that are attached to the main dwelling and under a common roof with the main dwelling. Interior pressurization of a structure occurs only when a hole in the building envelope reaches a critical size. As discussed below, most windows and all doors are of sufficient size that, if blown out or sucked in by wind pressure, the elimination of the entire unit creates a large enough hole for interior pressurization to take place. If a window or door is penetrated by a missile, though, the hole corresponding in size to the face of the missile impacting the building component may not be large enough to allow interior pressurization to take place. Wind-borne debris is a greater concern for windows that are not impact resistant because, after a window shatters, any hole typically will be the size of the unit of glazing struck by the missile. This case involves residential structures. As noted above,15/ Form 1699 divides homes into two categories: those that were initially permitted prior to the effective date of the FBC of 200116/ (existing) and those that were initially permitted under the FBC (new). As provided by section 627.0629(1), homeowners' insurers are required to provide a discount on the wind portion of an annual homeowners' insurance premium for the installation or implementation of the fixtures and construction techniques listed in the statute. Insurers may use the discounts identified in Form 1699, or they may calculate their own discounts, although, if they do so, rule 69O-170.017 requires the insurers to base their discounts on appropriate studies and data. All but two homeowners' insurers17/ have set their rates using the discounts in Form 1699. Due to the language, "Windows or All," Form 1699, on its face, does not provide a discount for fixtures or construction techniques applicable to doors, only windows. In other words, Form 1699 provides the same discount for enhanced windows as it does for enhanced windows and enhanced doors and provides no discounts for enhanced doors absent enhanced windows. Not evident on the face of Form 1699 is that the form provides discounts only for enhancements to impact resistivity, not wind resistivity. The dispute in this case arose because Petitioner manufactures a fixture to increase the wind resistivity, but not impact resistivity, of existing garage doors. "Opening protection" has two meanings. In the broader and more natural sense, "opening protection" means that all openings--i.e., windows and doors--are protected against both forms of wind peril--i.e., wind pressure and wind-borne debris. At times, the broader definition of opening protection has been used by OIR18/ and ARA.19/ More frequently, though, opening protection is limited to protection from wind-borne debris only and is thus a measure of impact resistivity.20/ Adopting this more restrictive definition, section 627.0629(1) distinguishes between opening protection, on the one hand, and window, door, and skylight strength, on the other hand. This Final Order therefore uses opening protection in its more restricted sense--i.e., the protection of windows and doors from wind-borne debris only. Many of the findings in this Final Order are to clarify the opacity that has attached to the use of opening protection by ARA and OIR; to reestablish the statutory distinction between opening protection and window, door, and skylight strength; and to assess the consequence of the omission from Form 1699 of any discounts for window, door, and skylight strength. Resolution of these issues requires persistent differentiation among the losses that may arise from the damage that wind may cause to doors, the damage that wind may cause to windows, the damage that wind-borne debris may cause to doors, and the damage that wind-borne debris may cause to windows. Generally, windows fail due to impact by wind-borne debris more often than wind pressure,21/ but doors, especially garage doors, fail almost exclusively from wind pressure, not wind-borne debris.22/ Resolution of these issues requires persistent differentiation between wind resistivity and impact resistivity- i.e., the ability to withstand damage, respectively, from wind pressure and wind-borne debris. Wind pressure testing subjects a building component to uniform static wind pressure to determine if the component can withstand a specified wind load. Wind-borne debris testing subjects a building component to two tests: first, an impact from a missile of a specified weight traveling at a specified speed and, second, cyclic pressurization to determine if, after impact, the component can withstand short bursts of specified wind loads. As used in this Final Order, impact testing includes both of these tests. Lastly, as is implied by the language of section 627.0629(1), "strong" and "weak" should not be applied to describe the impact resistivity of windows and doors--i.e., opening protection. The statute speaks of enhancing the strength of four features, including windows, doors, and skylights, but only of enhancing opening protection. The failure of the statute's third sentence to modify "opening protection" with "strength" is intentional and reinforces the distinction between opening protection, which is measured by impact resistivity, and the remaining mitigative features, including window and door strength, which are measured by wind resistivity. As discussed below, the FBC maintains the same distinction by defining "strong" and "weak" in terms of the relative ability of a building component to support a factored load.23/ Therefore, a "weak" or "strong" window or door refers to the resistivity of the window or door to uniform static wind pressure and possibly cyclic pressurization, but not to missile impacts. For these reasons, this Final Order refers to a "strong" or "weak" building or building component or "strengthening" a building or building component in terms of the ability to resist wind pressure, including cyclic pressurization, but not missile impact. Petitioner and its Secure Door® Bracing System A limited liability company located in Plantation, Florida, Petitioner manufactures and distributes the Secure Door® residential garage door bracing system. Invented in the mid 1990s and marketed since the late 1990s, the Secure Door® bracing system consists of telescoping aluminum struts that can extend to eight feet and are anchored in the header above the garage door and the floor below the garage door. Each strut attaches to the sectional hinges of a garage door. The width of the garage door dictates the number of struts in the bracing system. A single-car garage, which is eight to nine feet wide, requires only one strut. A two-car garage, which is 16 feet wide, requires two struts. Garage doors are typically held in place by two tracks at either end that guide the doors when raised or lowered. Garage doors in existing construction typically offer little resistance to wind pressure. As little as eight pounds per square foot of wind pressure is sufficient to force such doors to bow and pull off their tracks. Failed garage doors obviously leave a large opening unprotected from the wind--an especially dangerous situation when high winds rush into garages and pressurize the interior of the house. The Secure Door® bracing system is effective. It strengthens weak garage doors. A two-strut Secure Door® bracing system raises the wind resistivity of a garage door to about 31 pounds per square foot, and a three-strut Secure Door® bracing system raises the wind resistivity of a garage door to about 50 pounds per square foot. The Secure Door® bracing system is inexpensive. Online, a one-strut system sells for about $150, and a two-strut system sells for double this amount. Petitioner's bracing system is easy enough for many homeowners to install themselves, so any installation costs are small. The installed cost of Petitioner's bracing system is thus far less than the typical cost of $850 to $1500 for a new one-car or two-car garage door that complies with the wind resistivity requirements of the FBC. An approved independent testing laboratory has certified that garage doors fitted with the Secure Door® bracing system meet the requirements of the 2010 FBC for wind resistivity, if installed onto a nonglazed, sectional garage door constructed of at least 25-gauge steel. The Secure Door® bracing system is the sole product cited as having FBC approval for strengthening garage doors in the Hurricane Retrofit Guide published by the Florida Division of Emergency Management.24/ Petitioner's product has achieve a unique prominence in the hardening or fortification after-market, as reflected, in part, by the fact that over 100 Lowe's outlets in Florida stock the Secure Door® bracing system. From 2005-11, Petitioner has sold to Florida purchasers, directly and through Lowes, from 760 to 10,363 units annually for total revenues of about $3.7 million or about $500,000 annually. Petitioner secured about 75% of these sales in 2006-08--after the period of numerous hurricanes during 2004-05. Given the effectiveness, low cost, and availability of Petitioner's product, as well as the abundance of existing Florida homes with weak garage doors, a homeowners' insurance discount for the installation of a Secure Door® bracing system would likely increase sales significantly. Petitioner filed a petition to initiate rulemaking to provide a premium discount for homeowners installing a Secure Door® bracing system. However, on May 4, 2012, OIR denied the petition on the ground that Petitioner lacked standing. Hurricane Andrew The landfall of Hurricane Andrew 20 years ago ultimately prompted two major changes in Florida's construction and homeowners' insurance industries. Future construction statewide would be governed by the FBC, and future rate filings of homeowners' insurers would provide an elaborate set of discounts for mitigative fixtures and construction techniques. These two legislative achievements marked the "grand bargain" struck between the homeowners' insurance and homebuilding industries.25/ Four months after Hurricane Andrew, FEMA published its "observations, recommendations, and technical guidance" in a report entitled Building Performance: Hurricane Andrew in Florida (December 21, 1992) (FEMA Andrew Report). According to the FEMA Andrew Report, Hurricane Andrew was, at landfall, a category-four storm that produced high winds and high storm surge. Although the flood damage was "minimal," probably due to the compactness and speed of the storm after landfall," the wind damage was "widespread."26/ Even so, only "in some areas" did Hurricane Andrew's land-based wind speeds exceed the design wind speeds in the South Florida Building Code, so "properly designed and constructed buildings should have experienced fewer storm- related damages when factors of safety required by the [South Florida Building] Code [SFBC] are taken into consideration."27/ In other words, Hurricane Andrew demonstrated the inadequacy of the standards in existing building codes to protect against even the wind speeds assumed in these codes. Focusing on particular causes of damage, the FEMA Andrew Report states that widespread catastrophic building failure was primarily due to "negative pressure and/or induced internal pressure overloading the building envelope."28/ The FEMA Andrew Report warns: A breach of a building's envelope (i.e., the system by which the building resists wind penetration) is particularly hazardous during wind storms. [A]dditional direct internal wind pressures combine with suction pressures on exterior faces, causing partial or complete blowouts of major structural systems such as walls and roofs. Double-car garage doors and entry doors especially should be held secure during wind storms.[29/] Reinforcing the relationship between major structural damage and failed windows and doors, the FEMA Andrew Report finds: The breaching of the building envelope by failure of openings (e.g., doors, windows) due to wind or debris impact was a significant factor in the damage of many buildings. This allowed an uncontrolled buildup of internal air pressure that resulted in further deterioration of the building's integrity.[30/] Focusing on garage doors, the FEMA Andrew Report describes damage from wind pressure, but not wind-borne debris: The failure of garage doors was determined to have promoted a great deal of damage to buildings. . . . [G]arage doors failed when the door deflection exceeded the amount allowed for in the manufacturer's design. . . . Loss of the doors resulted in an envelope breach and a sudden increase in internal pressures to the buildings.[31/] FEMA's analysis of the failure of entry doors likewise focuses on damage from wind pressure to the exclusion of damage from wind-borne debris.32/ By contrast, wind pressure and wind-borne debris both endanger windows and large sliding glass doors. The FEMA Andrew Report documents "failure from high wind pressures and debris impact," adding that "glazing left without storm protection was especially prone to penetration by airborne materials and failure due to the wind loads."33/ The relevant conclusions of the FEMA Andrew Report are: a) preserving the integrity of the building envelope limits damages and losses; b) protecting windows and doors preserves the integrity of the building envelope; c) wind and wind-borne debris damage windows and d) wind, but generally not wind-borne debris, damages doors. Response to Hurricane Andrew General The Legislature responded to Hurricane Andrew on multiple fronts.34/ In 1993, the Legislature enacted section 627.0629, Florida Statutes.35/ The narrow scope of the original statute is evidenced by the fact that it was then known as the "shutter discount rule."36/ In 1997, the state of Florida launched the Residential Construction Mitigation Program, which sponsored detailed inspections of over 2000 houses in selected coastal counties between 1998-2000. Through the program, the state collected valuable data about Florida's housing stock prior to the 2001 FBC in terms of construction features, mitigation fixtures and construction techniques, and expected loss-mitigation benefits.37/ In 1998-99, the Florida Windstorm Underwriting Association developed a plan for classifying buildings based on wind-risk features. Previously, the insurance industry had classified buildings based exclusively on fire-risk features, which were not necessarily the same as wind-risk features. Examples of wind-risk rating factors included roof shape (hip versus gable), method of attaching roof sheathing, presence or absence of a garage, and extent of protecting openings in the frame. As ARA described it, the resulting "rating factors . . . are synergistic amongst multiple features and not simply additive . . . because each element of the building envelope is vulnerable and, hence, combinations of mitigation items interact nonlinearly."38/ In 2000, the Legislature enacted important amendments to section 627.0629(1), rendering the statute substantially in its present form.39/ Florida Building Code The uneven performance of SFBC-compliant homes in Hurricane Andrew demonstrated the need to strengthen and improve what, up to that time, had been regarded as the "best hurricane code" in the United States.40/ After the adoption in 1993 of wide-scale requirements for improved roof systems and in 1994 of major structural and building component upgrades,41/ by 1997, Florida essentially operated under the 1997 Standard Building Code (SBC) with Florida amendments and the SFBC, according to FEMA in a report entitled Hurricane Charley in Florida (FEMA Charley Report).42/ These post-Andrew building code revisions culminated in the legislative adoption of the 2001 FBC.43/ Developed and maintained by the Florida Building Commission, the 2001 FBC generally superseded all local building codes for new construction permitted on or after March 1, 2002.44/ For ease of reference, this Final Order generally cites from the most recently adopted revision of the FBC, the 2010 FBC. Addressing wind resistivity, the 2010 FBC provides: "Buildings, structures and parts thereof shall be designed to withstand the minimum wind loads prescribed herein."45/ Winds are determined in accordance with ASCE 7.46/ These are the provisions that subject buildings and their components to wind loads and effectively subject buildings and their components to wind resistivity standards. The 2010 FBC includes a wind contour map for the state of Florida, which, for single-family residences, depicts ultimate design wind speeds from 150-180 mph in Broward, Dade, and Monroe counties down to 115-120 mph in the central part of the Panhandle, including Leon County.47/ In "many applications," the design wind pressure in the 2001 FBC--and, later, the 2010 FBC--was higher than the design wind pressures specified in previous building codes.48/ The 2010 FBC provides a table devoted exclusively for determining wind loads for garage doors and rolling doors.49/ The wind loads are based on the area of the door and the pitch of the roof, but the maximum positive or negative wind load for 170 mph winds is about 30 pounds per square foot, 150 mph winds is about 22 pounds per square foot, and 130 mph winds is about 18 pounds per square foot.50/ These design wind loads, when applied to a 16' x 7' garage door, mean that the door must be able to support the equivalent of the force produced by placing the door over a large hole and parking a small car on it. The FBC divides Florida into three regions51/: the High-Velocity Hurricane Zone (HVHZ), which is all of Dade and Broward counties52/; the Wind-Borne Debris Region (WBDR), which is most of coastal Florida53/; and the remainder of Florida.54/ In addition to the wind resistivities described in the preceding paragraphs, new construction must also meet impact resistivities. For new construction in the WBDR, windows must be impact resistant.55/ For new construction in the HVHZ, windows and doors must be impact resistant.56/ Because of the dual requirements for doors in the HVHZ and, presumably, the unlikelihood that, rather than merely replace the existing garage door, a homeowner would retrofit her garage door with Petitioner's product and, say, shutters to protect against wind- borne debris, Petitioner does not market its device in Dade and Broward counties. 2002 ARA Report Later in the same month that the 2001 FBC went into effect, ARA published the 2002 ARA Report. The purpose of the 2002 ARA Report was to estimate the reduction in homeowners' insurance hurricane losses from the installation or implementation of the fixtures and construction techniques identified in section 627.0629 in existing single-family residences and from the construction of new, FBC-compliant single-family residences.57/ The introduction to the 2002 ARA Report quickly reveals a couple of problems. ARA acknowledged that the scope of its study had to cover at least the wind resistive features identified in section 627.0629(1), but: a) ARA collapsed two statutory features--opening protection and window, door, and skylight strength--into a single feature that it labeled "opening protection of windows, doors, and skylights"58/ and b) ARA described all five features as "wind resistive,"59/ even though, as discussed below, ARA studied windows and doors in terms of their impact resistivity, not wind resistivity. Notwithstanding these problems, the modeling that ARA undertook for the 2002 ARA Report represents a remarkable achievement. Previously, catastrophe modeling, at least for hurricanes, was based exclusively on the analysis of insurance loss data.60/ Using its HurLoss 2.0® hurricane-loss model (HurLoss), whose components comprise an overall climatological model and a hurricane wind field model,61/ ARA projected insured losses based on principles of engineering, climate science, and meteorology, including the analysis of the interaction of a building with wind,62/ as well as the analysis of insurance loss data, which was also used to validate the modeling.63/ In general, for existing construction, ARA acquired descriptive data of the housing stock.64/ ARA conceptualized six single-family residences,65/ distributed them to 31 distinct points within mostly coastal Florida (with four of these points at locations subject to the maximum design wind speed under the 2001 FBC of 150 mph),66/ and analyzed each residence--in terms of structural damage and loss to the dwelling, contents, and additional living expenses, but not other structures67/--with specific sets of mitigative fixtures and construction techniques.68/ Using HurLoss, ARA subjected each of these residences to 300,000 years' worth of hurricanes projected to strike each of the 31 locations.69/ Through this process, ARA projected, by residence, wind loads, wind-borne debris impacts, building component resistance values, damage states, economic losses resulting from the predicted damage,70/ and, by the application of principles of insurance, including the assumption of a 2% deductible, insured losses. 71/ Ultimately, ARA calculated the average annual loss for a specifically configured residence by taking the residence's total losses over the 300,000 years of simulated hurricanes and dividing the total losses by 300,000. Dividing the resulting average annual loss by the value of the home produced a loss cost. ARA introduced the notion of relativity by comparing the loss cost of a specifically configured residence to the loss cost of the typical Florida residence.72/ A ratio below 1.00 for a specifically configured residence means that the insured losses for that home are projected to be relatively low, and a ratio above 1.0 means that the insured losses are projected to be relatively high. To define the relevant configuration of an existing residence, the 2002 ARA Report identifies seven primary rating factors73/ and four secondary rating factors.74/ Compared to the statutory list of six mitigative features, ARA added a couple of features--terrain and roof shape--and omitted only one--window, door, and skylight strength. The levels of opening protection are "none," "basic," or "hurricane." "Basic" protects against the impact of a 4.5- pound missile, and "hurricane" protects against the impact of a 9-pound missile.75/ Although not explicit, the primary rating factor for opening protection is limited to windows. The secondary rating factor for opening protection extends to doors, provided windows are also protected against wind-borne debris-- i.e., windows and doors.76/ Using the seven primary rating factors, the 2002 ARA Report generates two tables of loss relativities--one for open terrain and one for terrain that is not open.77/ By using the correct table and identifying the primary rating factors, one can calculate an existing residence's loss relativity, which may require further adjustment if any secondary rating factors apply. As noted above, the primary rating factor for opening protection is limited in two important respects: a) it covers windows only78/ and b) it measures impact resistivity, not wind resistivity.79/ Extending opening protection to doors, the secondary impact factor also measures impact resistivity, not wind resistivity.80/ Given the omission of window and door strength from its loss relativity study, ARA therefore assessed only those fixtures and construction techniques that increase the impact resistivity of windows and doors. Ultimately, ARA determined that the secondary rating factor for opening protection produced a negligible effect. The adjustment for the secondary rating factor for protecting windows and doors from wind-borne debris multiplies by .98 the loss relativity ratio derived from the primary rating factor for protecting only windows from wind-borne debris.81/ The effect of the findings in the preceding two paragraphs was that ARA transmitted to OIR no loss relativities on which OIR could calculate discounts for fixtures and construction techniques that increased the wind resistivity of windows or doors and negligible loss relativities, relative to those for fixtures and construction techniques that increased the impact resistivity of windows, on which OIR could calculate discounts for fixtures and construction techniques that increased the impact resistivity of doors. Although groundbreaking, the modeling in the 2002 ARA Report was not definitive for several reasons, as ARA itself acknowledged: The estimation of losses for buildings with specific engineering details is an emerging technology and has many limitations. . . . The HURLOSS computed relativities have been compressed using a judgment factor. The resulting loss relativities, while reasonable estimates at this time, are likely to evolve with more data and further model improvements. . . . the scope of the project was extremely complex and the schedule limited. Many pieces of the work were done in parallel and many simplifications were needed to produce a final product. There is clearly room for refinement and improvement and a strong need for more data.[82/] One of the "many simplifications"--unwitting or intended--was to reduce the six statutory mitigative features to five. An interesting question is whether this simplification was "needed to produce a final product."83/ The ARA engineer called by Respondents tried to justify ARA's omission of window, door, and skylight strength from its 2002 loss relativities study. The engineer testified that impact testing, which includes cyclic pressurization, includes wind pressure testing and implied that impact resistivity therefore subsumes wind resistivity.84/ This testimony was misleading and unworthy of any weight, although, in fairness to the engineer, this testimony may have been based on a statement that appeared, once, in the 2008 ARA Report.85/ Among the flaws in this argument is that ARA's recognition of enhanced wind resistivities only in the form of enhanced impact resistivities ignores measures that increase wind resistivity, but have no effect on impact resistivity. An example of such a measure would be reinforcing the attachment of a frame of a door or window to the frame of the house; the result would be to increase the door or window's wind resistivity, but not impact resistivity. If the only measure of increased wind resistivity is through increased impact resistivity, measures such as reinforcing frames, although effective mitigation, would never generate reduced loss relativities or discounts. Of course, a more pertinent example of a measure that increases wind resistivity, but not impact resistivity, is Petitioner's product. ARA's recognition of enhanced wind resistivities only through enhanced impact resistivities also ignores the distinction between opening protection and window, door, and skylight strength in section 627.0629(1) and the distinction between impact resistance and wind resistance in the FBC. By including opening protection and window, door, and skylight strength among the six required mitigative features, section 627.0629(1) ensures that all components installed in all framed openings--i.e., windows and doors--will be assessed in terms of both relevant perils--i.e., wind pressure and wind-borne debris. By claiming that impact resistivity subsumes wind resistivity, the ARA engineer destroys the logic of the statutory scheme and renders meaningless the "window, door, and skylight strength" clause. The ARA engineer's argument also fails to harmonize with the provisions of the FBC calling for the separate assessments of opening protection and window, door, and skylight strength. For homes in the HVHZ, the 2010 FBC requires that doors and windows, among other building components, undergo testing for wind pressure,86/ impact,87/ and cyclic pressurization.88/ Testing Application Standard (TAS) 202-94 specifies the protocols for wind pressure testing of building components, and TAS 203-94 specifies a different set of protocols for cyclic pressurization testing, after impact testing, of building components.89/ If the cyclic pressurization test within the impact test subsumed the wind pressure test, the FBC would not maintain these distinctions. In his deposition, the ARA engineer never said that cyclic pressurization testing subsumed wind pressure testing. At hearing, the ARA engineer seemed unwilling to stake his testimony entirely on this flawed argument. He later testified that ARA failed to calculate loss relativities for the wind resistivity of existing doors and strengthened doors for reasons of convenience, but justified this failure by claiming that ARA's response to the invitation to negotiate had described the scope of the work that it would undertake.90/ The ARA engineer estimated that the loss relativity of a wind resistive window would fall between the loss relativities of the first two levels of impact resistivity for a window: none and basic.91/ The engineer did not specify a strength for the enhanced window. Even if he had, it is entirely unclear how the ARA engineer determined that the loss relativity of a window with a strength of, say, 45 pounds per square foot92/ would produce a loss resistivity somewhere between--if such a point can even be determined--the loss relativities of a window incapable of withstanding the impact of a missile and a window capable of withstanding the impact of a 4.5-pound missile. The ARA engineer estimated that the loss relativity of a wind resistive door would fall between the loss relativities of unspecified levels of protection, in terms of impact resistivity, for windows and windows and doors.93/ Like the estimate for a wind resistive window, this estimate for a wind resistive door is rejected as speculation. The derivation in the preceding paragraph of a loss relativity for wind resistance from a loss relativity for impact resistance at least both involved windows, which are vulnerable to wind and wind-borne debris. But the derivation of a loss relativity for wind resistance from a loss relativity for impact resistance involving doors seems dubious on another ground: doors are not particularly vulnerable to wind-borne debris.94/ Both unpersuasive attempts to peg loss relativities for wind resistance to loss relativities for impact resistance attempt to mitigate ARA's failure to conduct the necessary study and produce loss relativities for windows and doors for wind resistivity. Given the synergistic interdependency of mitigative features when calculating loss relativities, as discussed below, these estimates of wind resistivities by the ARA engineer fail to persuade. The Administrative Law Judge gives far more weight, though, to the admission by the ARA engineer that ARA should "absolutely" conduct another study to "include wind pressures" for doors, windows, and vents.95/ Dr. Lawrence Twisdale, an ARA principal and author of the 2002 ARA Report and 2008 ARA Report, never suggested that cyclic pressurization testing subsumes wind pressure testing or that loss relativities for wind resistivity could be quickly derived from loss relativities for impact resistivity. During a brief telephone conversation with Petitioner's principal in 2009, Dr. Twisdale instead admitted the obvious--i.e., ARA had not studied garage doors that much and its studies had focused on windows. Providing loss relativities for the wind resistivity or strength of windows and doors would be inconvenient. Among other things, ARA would face the difficult tasks of reviewing and perhaps collecting data96/ and exercising professional judgment as to the wind resistivities of existing windows and doors whose strength would be enhanced by mitigative fixtures and construction techniques that would render them more wind resistive, perhaps to FBC standards. But these were the tasks that ARA performed as to other mitigative features and were the tasks that the procuring agency97/ intended to impose on ARA. Adding one or more rating factors for window and door strength would be burdensome. As ARA explained its decision not to add a separate rating factor for garage doors (although apparently only for their impact resistivity): "This approach simplifies the application of the relativities and the numbers of combinations required to be considered."98/ But simplicity, a virtue in 2002, is cast in a different light in the 2008 ARA Report: While the details of the proper interpretation and application of rate differentials contained herein may be viewed by some as overly complex, the procedures reflect both the inherent variations in construction techniques as well as the engineering details embedded in the FBC. Over simplification of the measures of loss reduction (e.g., by simplifying the results to smaller tables with fewer features) also seems counter to the state's investments in improving the building code as well as in promoting public awareness of mitigation techniques. Eliminating features for the purpose of achieving simplicity would also result in lost opportunities for both awareness and mitigation of those features (including code-plus construction). [A] 1% reduction in average annual residential loss in Florida would amount to an annual statewide loss reduction savings of about $50 million in 2008 dollars. Each 1% loss reduction that we can wring out of new construction (through FBC improvements) and existing construction (through mitigation) will result in long-term exponential reductions in statewide losses.[99/] In any event, the failure of ARA to generate loss relativities for fixtures and construction techniques that increase the wind resistivities of windows and doors left OIR with no means to calculate the discounts for such fixtures and construction techniques. The testimony of the ARA engineer also addressed ARA's calculation of the .98 loss relativity ratio for increasing the impact resistivity of doors. The ARA engineer testified that, in part, ARA determined this relatively small difference in loss relativities by comparing FBC-compliant garage doors to strong, but not FBC-compliant, garage doors with wind resistivities of about 20 pounds per square foot, instead of to weak garage doors with wind resistivities of about 10 pounds per square foot.100/ Ignoring for a moment the inherent problems of mixing wind and impact resistivities, by assuming that existing garage doors were relatively strong, ARA dampened the result of any fixture or construction technique that further increased the impact resistivity of such garage doors. If ARA had chosen a garage door with reduced impact resistance, the .98 loss relativity ratio would have been lower and perhaps not amenable to characterization as negligible. In any event, the failure of ARA to produce an accurate loss relativity for fixtures and construction techniques that increase the impact resistivity of doors left OIR with no means to calculate an accurate discount for such fixtures and construction techniques. Due to the flaws detailed above, Form 1699 lacks discounts for fixtures and construction techniques that enhance the wind resistivity of windows and doors and understates the discount for fixtures and construction techniques that enhance the impact resistivity of doors. But, due to the interdependency of loss relativities and discounts, these omissions and understatement also raise the real possibility of distortion among the other loss relativities and discounts. This interdependency is suggested by the above-cited statement in Form 1655: "[w]hen one discount is applied, other discounts are reduced until you reach your maximum discount." Discounts are not merely added until the maximum is reached. A second discount necessitates the reduction of the first discount, and so on, until the maximum discount is reached. But the interdependency is more complicated, at least when calculating damage states and loss relativities. As noted above, at least since the late 1990s, engineers have known that wind-risk factors, such a roof-to-wall attachment, opening protection, and window, door, and skylight strength operate synergistically, and these nonlinear features cannot merely be added because they combine nonlinearly.101/ In calculating loss reductions, ARA illustrates the nonadditive property of mitigative features with a simple example in the 2008 ARA Report. If a hip roof reduces losses by 20% when compared to a gable roof, an FBC roof reduces losses by 30% when compared to a nonFBC roof, shutters reduce losses by 40% when compared to windows without shutters (or other form of enhanced impact resistivity), and a one-story residence reduces losses by 30% when compared to a two-story residence, the total effect of these several mitigative fixtures and construction techniques cannot be a 120% reduction in losses.102/ Obviously, "the effects of mitigating one feature are different depending on the remaining construction/mitigation features."103/ ARA describes the nonlinear relationships among loss relativity factors by considering the varying effects of a secondary rating factor, depending on whether it is installed or implemented in a weak or strong house. The 2008 ARA Report explains: for a generally strong building with only one weakness, the effect of the secondary factor is generally many times that of the effect of the secondary factor on an average building. For a weak building the secondary factor may not be very important since the building has so many other problems.[104/] These are simple illustrations of the nonlinear relationships among mitigative features. Notwithstanding the smoothed results in ARA's loss relativities tables, the effect on loss relativities of a single mitigative fixture or construction technique varies, depending on whether the house is strong or weak--i.e., depending on the presence of other mitigative fixtures and construction techniques. The synergistic effect of multiple mitigative fixtures and construction techniques in calculating damage states is mentioned below.105/ Relying on the loss relativities in the 2002 ARA Report, on January 23, 2003, OIR issued Informational Memorandum OIR-03-001M, which advised homeowners' insurers that, to ensure that future rate filings comply with section 627.0629(1), they should replace the shutter discount with a more elaborate set of discounts based on the primary rating factors identified in the 2002 ARA Report. Because of what it characterized as an "economically [in]significant" relativity adjustment factor of .98 for enhanced impact resistivity for doors, especially considering the "already very detailed and complex Form 1699,"106/ OIR omitted any credit for enhancing the impact resistivity of doors. To discourage "large rate changes" in anticipation of the elaborate discount scheme that had replaced the shutter discount, OIR temporarily "tempered" the stated value of the discounts by half. In January 2003, OIR also issued a form entitled Suggested Wind Premium Credits for Existing Construction. This form is identical to Form 1699, except that the identified discounts are halved, in accordance with the tempering described in the preceding paragraph. Hurricane Charley On August 13, 2004, Hurricane Charley struck Florida just southwest of Punta Gorda. Like Hurricane Andrew, Hurricane Charley generated peak gust wind speeds greater than 120 mph,107/ so it served as a design wind event, now under the FBC, from landfall to 120 miles inland.108/ Wind speeds of more than 120 mph, which is the midpoint of Category 3 hurricanes, are the threshold at which the building envelope of existing construction is normally stressed.109/ Wind speeds in excess of 120 mph therefore are required "to validate the effectiveness of wind mitigation features, since at low windspeeds the losses are . . . other than [to] the structural envelope."110/ The FEMA Charley Report states that Hurricane Charley made landfall as a Category 4 hurricane with one-minute sustained wind speeds of 150 mph and, like Hurricane Andrew, produced little storm surge or coastal flooding due to its narrow size and speed.111/ New construction performed well in Hurricane Charley. FEMA found: "the structural systems of buildings designed and constructed to the 2001 FBC performed as expected and thus there was little to no damage to the structural systems of these buildings."112/ Existing construction, of course, suffered more extensive damage from the strong winds of Hurricane Charley. Focusing on the risk that wind poses to windows, FEMA found, among other things, that unprotected windows led to structural failures due to increased internal pressures. FEMA stressed the heightened vulnerability of windows to impact damage: "Windows can be broken by over-pressurization, but this damage is not as common as debris-induced damage."113/ FEMA found that "[g]arage doors [were] blown in or out, allowing wind inside garages and often causing significant structural damage to the garages."114/ FEMA attributed the failures of residential garage doors throughout the Port Charlotte and Punta Gorda areas to low resistance to wind, not impact from wind-borne debris: In some instances, the doors buckled and were pulled outward (suction failures). . . . In other instances, the doors were pushed inward (positive pressure failures) . . .. Many of the failures occurred because the doors had inadequate wind resistance[, as] . . . most of the double car garage doors in older homes were not high-wind or debris-impacted rated.[115/] Notwithstanding the observation that few existing garage doors were wind- or impact-rated, the FEMA Charley Report focuses exclusively on wind resistivity in its discussion of commercial garage doors.116/ FEMA quantified the size of the hole in an exterior wall, including framed openings, that is sufficient to breach the building envelope. Holes as little as one percent of the wall area affect internal pressures, and holes of five percent of the wall area cause equalization of internal and external pressures.117/ Petitioner's expert witness testified that the critical size of a hole is four square feet, meaning that interior pressurization can take place in holes of this size or greater in the building envelope.118/ Garage doors that have failed due to wind pressure typically satisfy either of these critical-size criteria.119/ From its observations of damage from Hurricane Charley, FEMA concluded: This damage indicates that insufficient attention has been given to selecting materials or components of the building envelope that will meet the building code requirements for wind and water resistance. [N]ot enough attention is paid to building envelopes.[120/] FEMA recommended that existing windows be retrofitted so as to be impact resistant and existing garage doors be retrofitted so as to be "wind and impact resistant"--or, addressing only wind resistance--a design professional specify bracing for the door and reinforcement for the track.121/ FEMA found that the resistance of new construction to wind damage could be shared by existing construction that has undergone properly performed mitigation.122/ The Institute for Business & Home Safety (IBHS) and Door Access and Access Systems Manufacturer's Association prepared a report entitled Hurricane Charley: Garage Door Permit Study. Especially important given the extensive garage door damage observed by FEMA, IBHS determined that fixtures and construction techniques that increased the wind resistivity of garage doors provided substantial mitigative value in high winds.123/ 2008 ARA Report, FEMA HAZUS Manual and Florida Division of Emergency Management Procured by OIR, the 2008 ARA Report, which ARA issued in October 2008, incorporated the damage and insurance loss data available from the 2004 and 2005 hurricanes to determine, again, loss relativities for certain fixtures and construction techniques, on which OIR could rely in calculating discounts for these fixtures and construction techniques.124/ Among the important differences from the 2002 study were the modeling of 500,000 years' worth of storms,125/ two-story and multifamily residences, and recent changes to the FBC. Although the 2008 ARA Report adds rating factors, it does not add a rating factor for the wind resistivity of windows and door, nor does it contain a restated loss resistivity for the impact resistivity of doors. Converting into insured losses the damage observations of FEMA following Hurricane Charley, the 2008 ARA Report states that 125 mph winds produced dwelling losses of only 5% of policy limits for new construction, but 25% of policy limits for existing construction. Assuming that none of the existing construction had undergone wind mitigation, this meant that FBC-compliant construction had reduced insured losses by about 80%.126/ Bearing out the role of progressively more wind resistant building codes, whose inception was about 1995, ARA determined that wind losses trended down, based on the year of construction, from 1995 onward.127/ Prospectively, the adoption of the FBC has largely addressed the risk posed to new construction by hurricane winds. Retrospectively, though, much remains to be done. Bearing out the critical role of wind mitigation of Florida's existing housing stock, the 2008 ARA Report states: a) the insured value, in terms of dwelling coverage only, of owner-occupied, one-to-four-unit residences in Florida was nearly $1.1 trillion; b) the share of the insured value of that existing housing stock constructed prior to 1995 was about 59%; b) the share of the insured value of that existing housing stock constructed prior to 2002 was about 78%; and c) the share of the insured value of that housing stock constructed after 2001- roughly corresponding to what this Final Order has characterized as new construction--was about 22%.128/ Although the ratio of new construction to existing construction will continue to rise, in 2008, the insured value, for dwelling coverage only, of one-to- four-unit, owner-occupied existing residential construction was about $850 billion (after allocating a proportionate share of the insured value of residences whose age was unknown). Mitigative fixtures and construction techniques can reduce this $850 billion exposure--potentially, by as much as 80%, as mentioned above. The 2008 ARA Report provides a measure of confirmation of the findings, repeated elsewhere in this record, that doors are uniquely vulnerable to wind, not wind-borne debris, while windows are more vulnerable to wind-borne debris, but are still highly vulnerable to wind. The following findings are illustrative, not definitive, because they report the outputs of modeling runs of two homes with merely two sets of mitigative fixtures and construction techniques--one set resulting in a weak house and the other set resulting in a strong house. Appendix E of the 2008 ARA Report provides individual building analysis reports for a weak residence (with a nonFBC shingle roof covering and no protection of windows against impacts, among other features) and a strong residence (with an FBC tile roof covering and protection of windows against impacts, among other features).129/ Both modeled homes, which have attached garages, were exposed to over 2500 Category-3 or higher hurricanes. For the weak house, despite exposure to wind speeds of up to 200 mph, no garage door ever failed due to missile impact, only pressurization.130/ For the strong house, again despite exposure to wind speeds of up to 200 mph, no garage doors failed--from impact or pressurization.131/ In the weak house, windows failed much more often from missiles than wind pressure up to about 150 mph,132/ and, in the strong house, windows failed much more often from missiles than wind pressure up to about 185 mph.133/ Another source of data concerning the performance of garage doors, FEMA's technical manual to the HAZUS®MH MR2 hurricane model states that the replacement of a solid wall, which provides the greatest protection, with a strong garage door,134/ results in a "negligible" increase in modeled damage.135/ But the replacement of a solid wall with a weak garage door136/ results in an average increase in damage ranging from 2-24%.137/ As noted above,138/ the conversion of damage states to losses is a complicated process, but this difference in damage states is at least consistent with the inference that ARA's loss relativity adjustment for increased impact resistivities in doors is too conservative. Addressing the interdependency of mitigative fixtures and construction techniques, the FEMA technical manual warns: "the building performance is governed by the performance of the weakest link in the chain. If there are two links of roughly equal weakness, both must be strengthened."139/ FEMA illustrated this principle by noting that the average reduction in damage for a particularly configured home was 15% for enhancement of window protection and 27% for enhancement of roof strength, but the combined total of these two mitigative fixtures or construction techniques was 52%, not 47%.140/ According to the Hurricane Retrofit Guide published in 2010 by the Florida Division of Emergency Management, the "greatest risk" posed by hurricanes for most houses is to roofs and then windows and doors.141/ For most existing structures, which lack hurricane clips or straps holding the roof to the walls, "window and door protection may make the critical difference between losing your roof and keeping it on."142/ The Hurricane Retrofit Guide reports that windows are more susceptible to damage from wind-borne debris than wind pressure. The Hurricane Retrofit Guide warns that shutters over windows or doors are effective against damage from wind-borne debris, but "probably won't keep the doors and windows from bursting open from wind pressure if they are weak or poorly anchored to the walls of the house." If a window can withstand 40-50 pounds per square foot of wind pressure, the Hurricane Retrofit Guide continues, the window will "have a chance of resisting hurricane wind pressures," but most windows installed before 1996 are not strong enough to resist hurricane wind pressures.143/ Noting the size of garage doors and the prevalence, among older homes, of garage doors with low resistance to wind, the Hurricane Retrofit Guide concludes that garage doors may pose a heightened risk to homes: Because garage doors are so big, usually the single biggest opening on your house, and because most old doors are so weak, the survival of your garage door is very important to the survival of your house. Experience has shown in older houses that when garage doors fail in hurricane winds a lot of additional damage follows. When you are evaluating your house for its vulnerability to hurricanes you need to look at your garage door as one of the first and most important vulnerabilities. . . . * * * Failure of the garage door allows the full fury of the hurricane to enter the house to act on interior walls, doors, ceiling or roof that form the barriers between the inside of the garage and the rest of the house. This frequently leads to failure of these surfaces and can lead to significant loss of roof sheathing or loss of a part of the roof.[144/]
The Issue The issues for disposition in this proceeding are whether, as alleged in the Administrative Complaints dated December 15, 2000, and April 3, 2001, Respondent committed negligence in the practice of structural engineering, and if those allegations occurred in violation of Section 471.033(1)(g), Florida Statutes, what appropriate penalty should be recommended.
Findings Of Fact Based upon my observation of the witnesses and their demeanor while testifying, the documentary evidence received and the entire record complied herein, the following relevant facts are found. At all times material hereto, Petitioner, Florida Engineers Management Corporation, was the state agency responsible for regulating the practice of engineering in this state. At all times material hereto, Respondent, Shields E. Clark, P.E., was licensed as a professional engineer in the State of Florida having been issued license No. 6826 in 1958. In the 43 years Clark has been licensed as a professional engineer, this proceeding is the only allegation of professional wrongdoing of disciplinary action taken against him. Philip Matonte is a self-employed builder/property manager and owner of William Michael & Associates, Inc. This business entity is owner of two residential properties. One property is located at 397 Leeward Drive, Clearwater, Florida, and the other property is located in Seminole, Florida. DOAH Case No. 01-2290PL - 276 Bayside Drive, Clearwater, Florida. Respondent was the engineer of record as reflected by his signature on two sheets of plans for proposed structural additions to the residential property owned by Mr. and Mrs. Miller. Both sheets, date stamped June 26, 2000, had the logo DCI, Design and Consultant's, Inc., Drafting Services. R. Brazil drew the plans and Respondent signed and sealed both sheets. On or about August 4, 2000, the above-referred two sheets of proposed structural additions to the Millers' residence were submitted to the City of Clearwater for a building permit and were rejected by the City. The basis for the reviewers' rejection were: plans were vague and missing considerable details pertaining to existing beams and were missing information pertaining to the foundation, load paths for additional weight, and anchoring for the foundational structure. Petitioner's expert, John Power, opined that the above drawings failed to provide sufficient information to indicate the nature of the existing construction or sufficient detail to permit an evaluation of what effect the new construction would have on the existing structure. According to Power, he could not determine whether the carport had a roof; whether the carport had walls; if so, the wall section was not provided; whether the roof would have to be removed to accommodate the additional construction; whether there was a beam or wall at the back of the carport; and no details provided for the double 2 by 12 beam at the rear of the proposed addition to the existing structure. Power opined that Respondent failed to provide necessary notes and specifications in the drawings. First, he failed to specify the concrete strength for the new pilaster at the rear of the building and failed to specify the reinforcement steel grade. Second, he failed to specify the required concrete steel ties and failed to indicate the framing of the proposed skylight. Respondent's explanation that he provided xeroxed copies of specific code sections pertaining to each changed section of the plans with instructions to the person drafting the plans to include the code section requirements provided is evasive. Power further opined that Respondent failed to provide necessary notes and specifications in his drawings. He failed to specify the reinforcement steel grade; failed to specify the required concrete steel ties; failed to indicate the framing of the proposed skylight; and failed to specify the grade and species of lumber, that should have been specified, for the rafters. Power reasoned that it is unlikely a builder would use anything of a higher-grade No. 1 Southern Pine lumber. Even so, grade No. 1 Southern Pine would have been inadequate and overstressed, given the proposed structural additions and the double 2 by 12 members forming the beam at the rear of the building. Respondent's drawings calling for an 8 by 16 concrete beam in the front of the building was deficient because the beam is overstressed in shear and lacks the required stirrups for attachment. Further, there are no details as to how the platforms cantilevered from the exterior walls in Section F3 of the drawings are to be supported. Included on the blueprint drawings, dated June 26, 2000, was the statement: "This structure has been designed to resist uplift pressures generated by a 102-MPH wind," with the number 102 crossed out and the number "130" written underneath. From this change, Power opined that Respondent's method of calculation for basic wind speed was not clarified nor indicated on the drawings. The wind speed calculations and the added weight of the proposed structural construction form the basis for determining the stress imposed upon the existing structure. Power and Respondent agree that there are two acceptable methods for calculating and determining wind speed. The 1997 Standard Building Code ASCE 7-95 determines a basic wind speed from anticipated three-second gusts. Conversely, the Standard Building Code 1606.2 determines a basic wind speed from anticipated fastest mph wind velocity within a given time period 10, 15, 20, 30 or 50 years. According to Respondent, had Power based his computation upon Standard Building Code 1606.2 standard of the fastest wind in a given yearly cycle instead of the anticipated three-second gusts, he could not conclude the supporting beams were overstressed. Respondent's use of the 1606.2 standard, though calculations are not shown in the plans, is an acceptable engineering practice. DOAH Case No. 01-2396PL - 397 Leeward Drive, Clearwater, Florida. According to Matonte, after the City of Clearwater recommended Respondent to him as an engineer who had provided acceptable plans in the past, he met with Respondent several times to discuss hiring him for a project on his home at 397 Leeward Drive, Clearwater, Florida. According to Tom Chaplinsky, City's Plan Reviewer, no one from the City recommended Respondent to Matonte. I find Chaplinsky's testimony credible. According to Respondent, he met and discussed with Matonte the drafting of a steel beam to support additions to a swimming pool on a property located in Seminole, Florida, not in Clearwater, Florida. Matonte drove Respondent to the Seminole property and Respondent completed, as requested, a survey and the property and drew the steel beam, charging Matonte $150.00 for the single sheet drawing. Matonte's testimony on the above points is not credible. First, the City of Clearwater did not refer Respondent to him for professional engineering services. Second, a company, not Matonte, owns the residence at 397 Leeward Drive, Clearwater, Florida. Third, Matonte's company, William Michael & Associates, Inc., does own residential property in Seminole, Florida. The three sheets of plans for the 397 Leeward Drive property contain the signature and seal of the Respondent. However, Matonte drew 26 sheets of plans that he signed and sealed in the presence of Tom Chaplinsky and from which the City filed a complaint against him for signing and sealing as a professional engineer when he was not so authorized. The record does not reflect the specific sheets of plans Matonte presented to the City which were subsequently reviewed and rejected by Olson, another reviewer. Respondent takes the position that he did not sign and seal plans for a Clearwater property. His explanation is that "someone else, probably Matonte, using his seal, signed and sealed blank sheets during one of the several visits to his office and later cut and paste structural drawing on those sheets". I find Respondent did sign and seal the three sheets of plans at issue. Power, in his review of those three sheets of plans, opined that those plans did not contain details to explain Respondent's use of the "130-MPH" wind uplift resistance. From this basic premise, Power's opinion on this issue is the same as found in paragraph No. 11 herein above. Power's conclusions, however, are slightly different from his conclusions on the 276 Bayside Drive, Clearwater, Florida, property. Here he concludes that were a builder to rely solely on the drawings, the resulting construction would constitute a threat to life and [public] safety. At first blush, it would appear Petitioner's above allegation has substance. Taken as a whole, in the opinion of Power, once he decided the 130-MPH wind resistance notation was made over the crossed-out 102-MPH notation, finding deficiencies was a matter of review when using computation based upon anticipated three-second gusts. To this proposition, Respondent correctly contended that if Power used the 130-MPH wind resistance computation based upon fastest wind velocity over a period of years, his results would show the structure to be sufficient. Upon consideration of the fact that both Respondent and Power each have over 40 years engineering experience; that Respondent has worked almost exclusively on one/two story dwelling in Pinellas County for the past 10 to 15 years using the 1606.2, Standard Building Code, for the 130-MPH computation without complaint, compared to Power's most recent engineering practice has been primarily that of an expert witness; Power's testimony on this point is less than convincing.
Recommendation Based upon the foregoing Findings of Fact and Conclusions of Law, it is, therefore, RECOMMENDED that a Final Order be entered finding that Respondent, Shields B. Clark, is guilty of negligence in the practice of engineering and violated Section 471.033(1)(g), Florida Statutes, as set forth in Counts I and II in DOAH Case No. 01-2290PL, finding that Shields E. Clark is not in violation of Section 471.033(1)(g) as set forth in Count III of DOAH Case No. 01-2290PL and Count I of DOAH Case No. 01-2396PL, and dismissing Count II of DOAH Case No. 01-2396PL. Further, giving Shields B. Clark a written reprimand, placing him on probation for two years under such terms and conditions as the Board deem appropriate, imposing a fine of $2,000.00 and assess no cost related to the investigation and prosecution of theses cases. DONE AND ENTERED this 19th day of December, 2001, in Tallahassee, Leon County, Florida. FRED L. BUCKINE Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 SUNCOM 278-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us Filed with the Clerk of the Division of Administrative Hearings this 19th day of December, 2001. COPIES FURNISHED: Brian A. Burden, Esquire Post Office Box 767 Tampa, Florida 33601-0767 Natalie A. Lowe, Executive Director Board of Professional Engineers Florida Engineers Management Corporation 2507 Callaway Road, Suite 200 Tallahassee, Florida 32303-5267 Douglas D. Sunshine, Esquire Florida Engineers Management Corporation 2507 Callaway Road, Suite 200 Tallahassee, Florida 32303-5267 Hardy L. Roberts, III, General Counsel Department of Business and Professional Regulation Northwood Centre 1940 North Monroe Street Tallahassee, Florida 32399-2202
The Issue Whether Respondent was negligent in the practice of engineering as a result of alleged deficiencies in a porch addition to a residential structure, as alleged in the Administrative Complaint.
Findings Of Fact Based upon the evidence, the following Findings of Fact are made: Petitioner is charged with regulating the practice of engineering pursuant to Chapters 455 and 471, Florida Statutes (2004). Respondent is, and has been at all times material to this matter, a licensed professional engineer in the State of Florida, having been issued License No. PE 12217. Pursuant to the terms of a stipulation entered into between Respondent and the Board of Professional Engineers (Board) on a prior matter, Respondent provided the Board with a list of all Respondent's projects completed at 6- and 18-month intervals. One of these projects was a Terks Contracting Construction, Inc. (Terks Contracting), project for a screened porch addition. Respondent testified he was retained for the purpose of producing drawings containing specifications for the contractor for the slab, supports, spanning members, the connectors, and other elements. Respondent is the engineer of record for the Terks Contracting's covered porch addition to an existing single- family house. Respondent analyzed the windloads for the wood joists, wood beam, and the uplift supports. He calculated not only the gravity load, but considered the uplift load, as well. The analysis of windloads are covered by Tables 1606.2A and 1606.2B of the FBC. Petitioner testified he used these tables in his design of the covered porch. These tables are a simplified version of the standards of the American Society of Civil Engineers, Chapter 7-98 (ASCE 7). Respondent's determination of the components and cladding wind pressure are not inconsistent with the standards of the ASCE 7, which is the windload provision of the FBC. Respondent's calculations state that the rafters experience a uniform uplift load of 34 pounds per linear foot (plf) for the interior zones, and 38.4 pounds per square foot (psf) and 30.1 psf, respectively, for the endzones. Respondent is not mistaken in treating the overhang area and the endzones as two different areas adjacent to each other. To the extent the endzone (or edge zone) has higher wind pressures, Respondent calculated that the new structure would not adversely affect the existing house, if the new joists were attached to the top plate of the existing house. Respondent's calculations and drawings do not adversely affect how the covered porch addition affects the existing structure. Respondent's plan calls for the new joists to be connected to the top plate, which is part of the wall of the existing house, using a Simpson H 2.5 connector and to be set along side the existing truss members. The new joists were not attached to the existing trusses. In addition, a nailing pattern was selected for the plywood sheathing panels that accounted for variable wind pressures using the higher endzone windload for the entire area. However, Aranegui testified that he did not make any calculations to determine if, in fact, the specifications in the drawings were inadequate or incorrect. Petitioner's witness testified that he did not inspect the structure or speak with Respondent regarding the design of the structure. Aranegui did not take issue with Respondent's specifications for components or materials for the structure. Petitioner presented no evidence that there had been any complaints about the integrity of the structure, that the structure was deficient or unsafe in any manner, or that the structural elements as shown on the plans and, as built, did not meet engineering or FBC requirements. Aranegui testified that his opinions were based solely on two documents: the plans for the additions, Petitioner's Exhibit 1; and Respondent's calculations, Petitioner's Exhibit 2. Petitioner's expert witness testified that his concern came from his review of the paper calculations done by Respondent, and he believed the calculation sheet (not the plans) did not show consideration of windloads or complete analysis. However, Aranequi did not identify any requirement that these calculations be shown on paper and not just on the plans, nor did he identify any engineering practice imposing a duty on Respondent to write these calculations down. As noted below, there is convincing evidence that these concerns were considered and addressed by Respondent in the plans. Aranegui testified that his concern arose from his belief that the entire project in question should have been treated as a large overhang. However, Aranegui did not identify any provisions of the building code requiring Respondent to treat the screen porch addition as an overhang and did not identify any specific engineering practice that imposed a duty on Respondent to approach the design of the project as a large overhang. Aranegui's opinions were based on the assumption that because the plans and the calculation sheet he reviewed did not, in his opinion, show that certain matters were considered, they were not, in fact, considered in the design of the structure. Aranegui also testified he did not find an actual error or miscalculation in Petitioner's Exhibit 2. The evidence is convincing that the design of the porch addition was sound and safe and met all required structural requirements. The evidence is not clear and convincing that Respondent was negligent in the practice of engineering in a porch addition to a residential structure.
Recommendation Based on the foregoing Findings of Facts and Conclusions of Law, it is RECOMMENDED that the Board of Professional Engineers enter a final order finding Respondent, James Tippens, P.E., not guilty of negligence in the practice of engineering as alleged and that the Administrative Complaint filed in this cause be dismissed. DONE AND ENTERED this 5th day of May, 2005, in Tallahassee, Leon County, Florida. S DANIEL M. KILBRIDE Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 SUNCOM 278-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us Filed with the Clerk of the Division of Administrative Hearings this 5th day of May, 2005. COPIES FURNISHED: Douglas D. Sunshine, Esquire Florida Engineers Management Corporation 2507 Callaway Road, Suite 200 Tallahassee, Florida 32303-5267 Jawdett I. Rubaii, Esquire 1358 South Missouri Avenue Clearwater, Florida 33756 Natalie A. Lowe, Executive Director Board of Professional Engineers Department of Business and Professional Regulation 2507 Callaway Road, Suite 200 Tallahassee, Florida 32303-5267 Leon Biegalski, General Counsel Department of Business and Professional Regulation Northwood Centre 1940 North Monroe Street Tallahassee, Florida 32399-2202
The Issue The issues in this case are whether each of the two respondents practiced contracting and electrical contracting without a license in violation of Subsections 489.113(2), 489.127(1)(f), 489.531(1), Florida Statutes (2004),1 and, if so, what penalty, if any, should be imposed pursuant to Subsections 455.228(1) and 489.13(3).
Findings Of Fact Petitioner is the state agency defined in Subsection 489.105(2) that is responsible for regulating the practice of contracting and electrical contracting pursuant to Subsection 455.228(1). Neither of the respondents has ever been licensed as either a contractor or an electrical contractor. On April 14, 2005, Mr. Adams and Bay Breeze Maintenance, LLC (Bay Breeze), practiced contracting and electrical contracting within the meaning of Subsections 489.105(3) and (6) and 489.505(9) and (12). Mr. Adams, as agent for Bay Breeze, submitted to Mr. Christopher King, as agent for Dome Flea Market in Venice, Florida, a written proposal to remodel part of the Dome Flea Market for a cost not to exceed $60,000. The proposed remodeling involved an upgrade of a snack bar into a grill and bar to be known as the Sawmill Grill. In relevant part, the proposed remodeling required performance of plumbing, carpentry, and electrical contracting, including the installation of electrical wiring and electrical fixtures. Between April 14 and May 20, 2005, Mr. King paid approximately $39,350 to the respondents for the proposed remodeling job. On April 14, 2005, Mr. King paid $8,000 and $1,500 by respective check numbers 7725 and 7726. On April 19, 2005, Mr. King paid $8,000 and $1,700 by respective check numbers 7730 and 7731. On May 3 and 20, 2005, Mr. King paid $5,150 and $14,000 by respective check numbers 7742 and 7770. The respondents never actually performed any remodeling work. Mr. Adams testified that Mr. King forged the written proposal and that neither Mr. Adams nor Bay Breeze submitted a proposal for the remodeling project. That testimony was neither credible nor persuasive. The financial harm to the public was substantial. Mr. Adams and Bay Breeze have not paid any restitution. Petitioner incurred investigative costs, excluding attorney fees and costs, in the amount of $844.07. The investigative costs are reasonable within the meaning of Subsection 489.13(3).
Recommendation Based upon the foregoing Findings of Fact and Conclusions of Law, it is RECOMMENDED that Petitioner enter a final order finding Mr. Adams and Bay Breeze guilty of committing the violations alleged in each Amended Administrative Complaint and imposing an aggregate administrative fine against Mr. Adams and Bay Breeze, collectively, in the amount of $10,000. DONE AND ENTERED this 12th day of March, 2007, in Tallahassee, Leon County, Florida. S DANIEL MANRY Administrative Law Judge Division of Administrative Hearings The DeSoto Building 1230 Apalachee Parkway Tallahassee, Florida 32399-3060 (850) 488-9675 SUNCOM 278-9675 Fax Filing (850) 921-6847 www.doah.state.fl.us Filed with the Clerk of the Division of Administrative Hearings this 12th day of March, 2007.
