Patent Issued for Method Of Modeling Roof Age Of A Structure (USPTO 10,268,691)

2019 MAY 07 (NewsRx) -- By a News Reporter-Staff News Editor at Insurance Daily News -- From Alexandria, Virginia, NewsRx journalists report that a patent by the inventors Emison, Joseph Tierney Masters (Asheville, NC); Tachovsky, Holly Ward (Austin, TX); White, Richard W. (Trumbull, CT), filed on April 27, 2015, was published online on May 6, 2019.

The patent’s assignee for patent number 10,268,691 is BuildFax (Asheville, North Carolina, United States).

News editors obtained the following quote from the background information supplied by the inventors: “Field of the Invention

“The present invention relates to techniques for modeling roof age of a structure, which techniques can be used in the homeowners insurance industry in providing more accurate insurance coverage and risk analysis. More particularly, the present invention relates to a computer-implemented method for modeling roof age of a structure based on a generalized linear model employing one or more variables that influence roof age determinations.

“Description of Related Art

“The cost of replacing a roof due to wind, hail, or other weather damage can be significant and depends on the type of materials being replaced. For example, the cost to professionally remove and replace asphalt shingles, the most common type of roofing material, can exceed $8,000 for a typical ranch style home. The cost to replace more expensive materials such as metal, tile, or slate can reach into the tens of thousands of dollars. Further, roof damage is present in 85-95% of wind-related insured property losses each year, according to the Insurance Institute for Business & Home Safety (IBHS), and loses from thunderstorms cost insurers $14.9 billion in 2012, according to the Insurance Information Institute.

“As a typical homeowners insurance annual premium is only the fraction of the cost of a roof replacement, replacing a roof can be an expensive proposition for insurance companies. Although damage from wind, rain, and hail are typically covered by insurance policies, many insurance companies are taking steps to mitigate their losses. Because older roofs may be considerably weaker and thus more prone to damage, some insurance companies will not underwrite a policy with roof coverage for a home with a roof age over a certain limit. Also, some companies will only reimburse a depreciated value for a roof if it exceeds a certain age, such as ten years. Further, damage due to normal aging and wear and tear is typically not covered under home insurance policies. Thus, accurate information on the age of a roof at the time an insurance contract is underwritten is of considerable value to insurance companies.

“Homeowners insurance companies have traditionally relied on the homeowner to provide roof age information at the time of underwriting. However, homeowner-supplied information on roof age is often based on inaccurate information or misrepresented (since homeowners have an incentive to underestimate roof age), and is not validated by insurance companies. Research by the present inventors has shown that more than two-thirds of all homeowner-supplied roof ages are underestimated by more than five years, and that more than twenty percent are underestimated by more than fifteen years (See Emison and Tachovsky, HomeownerSupplied Roof Age is Disastrously Wrong, Claim Journal, 2013).

“These underestimates can result in significant loses for insurance companies when underwriting or managing a homeowners insurance policy, paying claims, or determining premiums. Thus, there is a need in the insurance industry for more accurate methods for estimating roof age of a property. To this end, methods of the invention aim to solve this business challenge by providing a more realistic roof age of a property based on a per-property modeled roof age.”

As a supplement to the background information on this patent, NewsRx correspondents also obtained the inventors’ summary information for this patent: “According to embodiments of the invention, a computer-implemented method for estimating roof age of a structure is provided. When referring to roof age of a structure, in the context of this specification this can refer to the year the structure was built, or the year the structure was built plus or minus a number of years (respectively to account for conditions that would either age a roof prematurely or add life to a roof), or roof age can refer to the actual age of the structure or roof (which can be determined by simple addition or subtraction of the roof year from the present year or some other point in time. For example, a roof of a structure built in 2001 could be said to have had a roof age of 10 years as of 2011. Since roof age can be determined easily from a year, the terms roof age and roof year may be used interchangeably in this specification.

“The method comprises identifying a target structure for which an estimated roof age is desired, receiving data related to the target structure (e.g., in computer-readable form), and inputting the data related to the target structure into a generalized linear model to determine a Per-Property Modeled Roof Age of the target structure. The modeled roof age preferably accounts for (i) the height of the trees, (ii) the proximity of the trees to the target structure, and/or (iii) the weather conditions for the structure’s geographic location. Such methods can be applied in practice by substituting a typically inaccurate reported roof age with a more realistic modeled roof age in making homeowners insurance decisions.

“Provided in embodiments, for example, is a method for determining per-property modeled roof age, the method comprising:

“(a) identifying a target structure for which a modeled roof age is desired;

“(b) obtaining one or more digital images of the target structure within a selected geographical radius using one or more satellite or aerial imaging apparatus;

“© identifying one or more tree within the selected radius from one or more signals represented in the digital image;

“(d) determining from the signal, proximity of the tree to the target structure;

“(e) converting intensity of the signal into height of the tree;

“(f) identifying frequency of weather conditions for a geographic area in which the target structure is physically located;

“(g) creating a mathematical model of roof age for the target structure based on:

“(1) the height of the trees,

“(2) the proximity of the trees to the target structure, and

“(3) the frequency of weather conditions for the geographic area; and

“calculating a modeled roof age of the target structure using the model.

“Such methods can comprise using the modeled roof age as a substitute for and/or as a more accurate alternative for reported roof age of the structure.

“Other data related to the target structure may be used including age of the target structure (based on year built), type or rating of roofing material, number of stories, estimated height of home, one or more tree proximity measures, one or more tree height measures, and one or more weather factors in a geographic area of the target structure. The one or more weather factors may include frequency of local severe weather that threatens roofs, average wind speed in the area, record wind speed in the area, and average and record size of hail in the area. In embodiments, an Area Average Roof Age based on the data related to a plurality of structures in the vicinity of the target structure may be calculated according to methods described herein, and inputted into the generalized linear model to determine the Per-Property Modeled Roof Age of the structure. The tree proximity measure may be one or more tree proximity categories described herein.

“Also included is a method (e.g., a computer-implemented method) for determining per-property modeled roof age, the method comprising: identifying a target structure for which a modeled roof age is desired; identifying a number of trees within a selected radius of the target structure; measuring tree height and proximity of the trees to the target structure; identifying weather conditions for a geographic area in which the target structure is physically located; creating a mathematical model of roof age for the target structure based on: (i) the height of the trees, (ii) the proximity of the trees to the target structure, and (iii) the weather conditions for the geographic area; calculating a modeled roof age of the target structure (e.g., on a computer comprising a central processing unit (CPU)) using the mathematical model; and using the modeled roof age as a substitute for reported roof age of the structure.

“Such methods can comprise assigning the trees to a category defined by: (i) tall trees with a height of 20 feet and taller and located within a first radius surrounding the target structure; (ii) tall trees with a height of 20 feet and taller and located within a second radius surrounding the target structure that is larger than the first radius; (iii) medium trees with a height ranging from above 0 feet up to 20 feet and located within the first radius of the target structure; or (iv) medium trees with a height ranging from above 0 feet up to 20 feet and located within the second radius of the target structure.

“When assigning the trees to a category as outlined above, for example, each category can be characterized as a ‘Positive Tree Proximity’ if the trees in that category are capable of protecting structures in response to a particular weather condition or is characterized as a ‘Negative Tree Proximity’ if the trees in that category are capable of damaging structures in response to a particular weather condition.

“Additionally, according to embodiments, the mathematical model can be further based on: (i) age of the target structure, (ii) age of one or more structures within a radius of the target structure, (iii) type or rating of roofing material of the target structure, and/or (iv) number of stories or height of the target structure.

“In embodiments, for example, the age of the target structure can be obtained from Census, County Tax Assessors, state Building Departments, or real estate databases. Alternatively or in addition, in embodiments, the type of roofing material of a target structure can be obtained from purchase orders or invoices from roofing material manufacturers or roofing material distributors, from home inspections, or from purchase orders from roofing contractors.

“Alternatively or in addition, according to embodiments, the weather conditions can include one or more of: (i) frequency of weather in the geographic area capable of damaging roofs, (ii) average wind speed, (iii) record wind speed, (iv) average size of hail, (v) record size of hail, and/or (vi) intensity of weather-related fire incidents.

“The methods can further comprise calculating an area average roof age based on data obtained from a plurality of structures in the geographical area in which the structure is located and using the area average roof age in the mathematical model.

“Embodiments of the invention also include a method (e.g., a computer-implemented method) for determining per-property modeled roof age, the method comprising: identifying a target structure for which a modeled roof age is desired; identifying a number of trees within a selected radius of the target structure; measuring tree height and proximity of the trees to the target structure using vegetation density values obtained from satellite or aerial imagery; identifying frequency of weather conditions for a geographic area in which the target structure is physically located; creating a mathematical model of roof age for the target structure based on: (i) the height of the trees, (ii) the proximity of the trees to the target structure, and (iii) the frequency of weather conditions for the geographic area; calculating a modeled roof age of the target structure (e.g., on a computer comprising a central processing unit (CPU)) using the mathematical model; and using the modeled roof age as a substitute for reported roof age of the structure.

“In embodiments, the tree height and proximity of the trees to the target structure can be obtained from vegetation density values. For example, the vegetation density values can be obtained from satellite or aerial imagery. Such satellite or aerial imagery can be NAIP imagery in embodiments. Further, for example, vegetation density values can be Normalized Difference Vegetation Index (NDVI) values.

“Methods of the invention can use a mathematical model that is a generalized linear model, such as a Poisson regression.

“In particular embodiments of any of the methods described in this specification, the mathematical model can be defined by the following equation: Estimated roof year or age=[A*(Structure Age or Year Structure Built)]+[B*(Frequency of Weather in the Geographic Area Capable of Damaging Roofs)]+[C*(Positive Tree Proximity)]+[D*(Negative Tree Proximity)].

“In other particular embodiments of methods described in this specification, the mathematical model can be defined by the following equation: Estimated roof age=[A1*(Structure Age Between 0-15 yrs.)]+[A2*(Structure Age Between 16-30 yrs.)]+[A3*(Structure Age 31 yrs. or later)]+[B*(Frequency of Weather in the Geographic Area Capable of Damaging Roofs)]+[C*(Positive Tree Proximity)]+[D*(Negative Tree Proximity)].

“In the equations for estimating roof age of a structure, optionally the roofing materials can be integrated as additional variables, where a different adjustment for each type and amount of roofing material is taken into account. For example, the equations can include the following variables:

“To account for the material type of a roof, one or more of the following elements can be included in the equations: [E*(% of Roof that is Asphalt Shingle)]+[F*(% of Roof that is Tile)]+[G*(% of Roof that is Metal)]+[H*(% of Roof that is Slate)]+[I*(% of Roof that is Membrane]+[J*(% of Roof that is Cedar Shingle)], etc.

“Various aspects of these embodiments and other embodiments will be set forth in the drawings and detailed description.”

The claims supplied by the inventors are:

“The invention claimed is:

“1. A method for determining per-property modeled roof age, the method comprising: identifying a target structure; obtaining one or more digital images of the target structure within a selected geographical radius using one or more satellite or aerial imaging apparatus; identifying a number of one or more trees within the selected geographical radius from one or more signals represented in the one or more digital images; determining from the signal, proximity of the trees to the target structure; converting, by a computer processor, intensity of the signal into height of the trees; identifying type and frequency of weather conditions for a geographic area in which the target structure is physically located; determining a modeled roof age for the target structure as a substitute for reported roof age of the target structure using a generalized linear model based on: (a) the following Positive Tree Proximity factors providing protection to the roof of the target structure or Negative Tree proximity factors presenting a hazard: (i) trees having a height within a first height range and located within a first radius surrounding the target structure; (ii) trees having a height within the first height range and located within a second radius surrounding the target structure that is larger than the first radius; (iii) trees having a height within a second height range and located within a first radius surrounding the target structure, wherein the second height range is smaller than the first height range and does not overlap; and (iv) trees having a height within the second height range and located within the second radius surrounding the target structure; and (b) the frequency of weather conditions for the geographic area; and displaying the modeled roof age of the target structure.

“2. The method of claim 1, wherein the Positive and Negative Tree Proximity factors are further defined as: (i) tall trees with a height of 20 feet and taller and located within a first radius surrounding the target structure; (ii) tall trees with a height of 20 feet and taller and located within a second radius surrounding the target structure that is larger than the first radius; (iii) medium trees with a height ranging from above 0 feet up to 20 feet and located within the first radius of the target structure; or (iv) medium trees with a height ranging from above 0 feet up to 20 feet and located within the second radius of the target structure.

“3. The method of claim 1, wherein the generalized linear model is further based on: (i) age of the target structure, (ii) age of one or more structures within a radius of the target structure, (iii) type or rating of roofing material of the target structure, and/or (iv) number of stories or height of the target structure.

“4. The method of claim 3, wherein the age of the target structure or the age of the structures within a radius of the target structure is obtained from Census, County Tax Assessors, state Building Departments, or real estate databases.

“5. The method of claim 3, wherein the type of roofing material is obtained from purchase orders or invoices from roofing material manufacturers or roofing material distributors, from home inspections, or from purchase orders from roofing contractors.

“6. The method of claim 1, wherein the weather conditions include one or more of: (i) frequency of weather in the geographic area capable of damaging roofs, (ii) average wind speed, (iii) record wind speed, (iv) average size of hail, (v) record size of hail, and/or (vi) intensity of weather-related fire incidents.

“7. The method of claim 1, further comprising calculating an area average roof age based on a plurality of structures in the geographical area in which the target structure is located and using the area average roof age in the generalized linear model.

“8. The method of claim 1, wherein the generalized linear model is a Poisson regression.

“9. The method of claim 1, wherein the generalized linear model is defined by the following equation: modeled roof age, expressed as a year=[A*(Year Structure was Built)]+[B*(Frequency of Weather in the Geographic Area Capable of Damaging Roofs)]+[C*(Positive Tree Proximity)]+[D*(Negative Tree Proximity)], wherein: A, B, C, and D are each independently a number between -30 and +30; and the Frequency of Weather, the Positive Tree Proximity, and the Negative Tree Proximity is each independently selected from a number between 0 and 1.

“10. The method of claim 1, wherein the generalized linear model is defined by the following equation: modeled roof age=[A1*(Structure Age Between 0-15 yrs.)]+[A2*(Structure Age Between 16-30 yrs.)]+[A3*(Structure Age 31 yrs. or later)]+[B*(Frequency of Weather in the Geographic Area Capable of Damaging Roofs)]+[C*(Positive Tree Proximity)]+[D*(Negative Tree Proximity)]; wherein: A1, A2, and A3 are each a number between 0 and 30; B, C, and D are each independently a number between -30 and +30; and the Frequency of Weather, the Positive Tree Proximity, and the Negative Tree Proximity is each independently selected from a number between 0 and 1.”

For additional information on this patent, see: Emison, Joseph Tierney Masters; Tachovsky, Holly Ward; White, Richard W. Method Of Modeling Roof Age Of A Structure. U.S. Patent Number 10,268,691, filed April 27, 2015, and published online on May 6, 2019. Patent URL: http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=10,268,691.PN.&OS=PN/10,268,691RS=PN/10,268,691

(Our reports deliver fact-based news of research and discoveries from around the world.)