Joseph J. Chadwick and Joyce T. MeJunkin x
Facade Maintenance: Owner's Techniques for Data Management
Reference: Chadwick, J. J. and McJunkin, J. T., "Facade Maintenance: Owners' Techniques for Data Management," Building Facade Maintenance, Repair, and Inspection, ASTMSTP1444, J. L. Erdly and T. A. Schwartz, Eds., ASTM International, West Conshohocken, PA, 2004.
Abstract: The long term success of any building maintenance program is directly influenced by the information available. Facades are particularly challenging since they are not conventionally viewed as regular maintenance items and the components are mistakenly thought of as "permanent." The Standard Practice for Building Data
Management ASTM E-2166, gives the building owner a structure on which to arrange all relevant building data to develop factual relationships that will make forecasting, analysis, and budgeting for building maintenance more reliable.
Keywords: facade maintenance, building data, UN]FORMAT, XML, DTD Introduction
At twenty, you have the face you were born with.
At fifly, you have the face you deserve.
For the average building owner, this observation attributed to Coco Chanel presents a stingingly clear analogy to fagade maintenance. While personal care by most definitions is an individual exercise, building maintenance involves a complex orchestration of specialists over time. It is an orchestration where the conductor, musicians, and even the program changes with no predictable basis. In the face of executive transience, and unfortunately persistent personnel migration, the possibility of available and reliable fragments of building history to support immediate decision-making is becoming increasingly remote.
Few large buildings today are constructed as a testament to the business prowess or sentimentalities o f the patriarch owner. Although buildings have always been a tool of business, they had somehow lost the burden of cultural icon just as business had lost its interest in the tools of physical production. Consistent with the devolving business proclivities, the Tax Reform Act of 1976 defined Component Depreciation, which recast building consciousness in the business community in terms of"depreciable asset" a term once reserved for process equipment. In short, buildings were emphasized to be just another commodity in the service of the ubiquitous bottom line. Curiously, the guidelines for depreciating the "envelope" component represented a somewhat idealized lifespan of 1 Architect, Contract Administration and Coordinator, University Planning, respectively, Office of Facilities, Yale University, PO Box 208297, New Haven, CT 06520-8297.
109 Copyright* 2004 by ASTM International www.astm.org
1 10 BUILDING FAQADE, MAINTENANCE, REPAIR, AND INSPECTION
that entity - that is, the envelope was indexed to what was thought to be the life of a premium roof, since exterior walls were not understood to "wear out." Funding
frequently follows accounting conventions, rather than the actual rate of consumption or wear. Because budgets are typically set to satisfy screaming needs with funds available, it follows that decisions based on accounting rules based on misconceptions of physical science will not adequately support a realistic maintenance program. While it is always easy to blame the accountants, the root problem may lie with the lack of auditable information.
Building facades present unique challenges, even for the alert building owner. The conditions that can precipitate enormous problems are practically speaking, out of sight, and therefore out-of-mind. By the time a presentation of observable distress occurs, concealed damage is significant. Minor leaks that may have been dismissed to an overall nuisance-factor are actively ignored for years. Despite the implied lesson of useful life inherent in the now defunct component depreciation mechanism, money applied to maintaining a building's exterior is not clearly appreciated as having an obvious
"payback." New lobby finishes justify a raise in rent. New fire alarms are required by code and insurance. Replacing sealant at the end of its anticipated life definitely lacks glamour. All too frequently, action occurs only as a reaction to prominent failure, particularly in jurisdictions lacking a fagade inspection law.
The need for a coherent system to manage building data is enormous, as are the benefits. Historical data, lessons leamed, the locations of short-term fixes should not be lost with the retirement of an employee. Engineering observations can provide a critical benchmark - i f the report can be found. The ability to cross reference and inter-relate information can make possible more reliable identification of trends for prediction as well as more confident diagnoses. The economic interdependence of systems can be clearly demonstrated to make compelling cases to budget repair or replacement actions. The sum of many ill-fitting windows affects energy consumption; sealant failures can damage primary and secondary structure before interior finishes are affected. These concepts are certainly not new. Many consultants offer building management database services that, for commercial purposes, are specifically designed not to communicate outside of their proprietary system. The view of the data that is available is generally structured by a programmer with accounting advice designed for "high level decision makers." While the executive summary is unarguably necessary, God and the Devil do battle in the details. It is the day-to-day management and individual attention to the viscera o f the building that directly contributes to the long-term condition. The problem then is to make building data accessible and available as a useable tool throughout the chain of custody responsible for its upkeep. One should easily be able to research and analyze the history of conditions, cost, or actions taken on any system within a single building, and compare a subject building within a portfolio of buildings, or with buildings of similar use or construction types to a component level of detail.
Two significant deficiencies had stifled this effort - a standardized data structure and a sufficiently robust platform. The Standard Practice for Organizing and Managing Building Data, ASTM E-2166, addresses the problem of data structure. It marries two widely used and accepted organizing systems: UNIFORMAT II ASTM E-1557 and Master Format. Although many popular proprietary estimating systems successfully link the two systems, they have been assembled for the convenience of the practical contemporary
CHADWICK AND MCJUNKIN ON FẴADE MAINTENANCE 1 11
estimator and lack the flexibility and definitional precision necessary for data collection, especially in novel or archaic construction types.
tmIFORMAT II reduces a building to essential systems and subordinate fimctional elements within a three level hierarchy. The systemic and functional relationships are essentially a conceptual model of a building such that a technically inexperienced person can readily understand the basic relationships. The current UNIFORMAT II as a standard is a classification system that evolved from the first elemental classification attributed to the British Ministry of Education following the post WWlI school expansion program. This methodology was applied to construction programs in other British Commonwealth countries, such as Canada, and then the United States in the early 1970s. In 1973 the American Institute of Architects undertook to develop an elemental cost estimating format called MASTERCOST. In conjunction with the General Services Administration, a consensus format named UNIFORMAT was produced. Although it was not an official national standard, it did form the basis for any subsequent elemental formats in the United States. In 1989, the American Society for Testing and Materials Subcommittee E06.81 on Building Economics appointed a task group to develop a UN]FORMAT standard.
In 1992, the National Institute of Standards and Technology (N/ST) issued a special publication [1], in which the name UNIFORMAT :I was selected to emphasize that it is an elemental classification system similar to the original UNIFORMAT. The improvements made it more comprehensive, particularly with respect to mechanical systems and sitework. In 1992, the Construction Specifications Institute (CSI) issued an interim edition of UNIFORMAT based on the work in progress of ASTM. CSI also published a practice entitled AFF/180-Preliminary Project Descriptions and Outline Specifications,@
which recommended the use of an elemental project description (specification) based on UNIFORMAT at the schematic design phase. The objective of the classification format was to improve communication and coordination among all parties involved in a project, particularly between the design team and the client. The ASTM standard was approved in 1993 and designated: Standard Classification for Building Elements and Related
Sitework - UNIFORMAT II ASTM E-1557-93. ,In 1996, revisions were made providing a distinctive alphanumeric format for the elements similar to that incorporated by CSI in 1992.
By virtue of its hierarchical structure, UNIFORMAT II offered an organization to logically summarize functional elements. These elements inherently possess details of interest. A method to amplify and focus further on these details in an orderly coherent way is mandatory for the viability of any data structure. The hierarchy should be sufficiently deep to describe the artifact adequately without spiraling into the kind of hair-splitting that yields a burdensome number of levels and sub-sets that dead end in arbitrarily finite categories. Some elemental classification systems pursued ways to wring a narrower focus by creating as many as 12 subordinate tiers. While this exercise may have had some situational merit, the practical application of a 12-tier system with related categories to conventional databases could be numbing. The capacity of such a system to adapt to new or innovative types of construction was limited to the classifier's ability to force the aberrant piece of construction into an existing category. The exercise of forcing an evolution of functional systems to building parts is further rendered futile in the face of an existing, widely accepted classification system for materials.
Master Format groups materials and products into 16 Divisions relative to similarity
112 BUILDING FACADE, MAINTENANCE, REPAIR, AND INSPECTION
of material and function, and organizes the Divisions nominally in the order in which that material or product appears in the construction process. For instance, Division 01 is General Requirements, Division 02 is Sitework, Division 03 is Concrete, Division 04 is Masonry, etc. All materials and products are coded with a five digit number, with the leading two digits representing the Division (Concrete: 03120), the next digit designating the broad-scope category (Concrete Formwork: 03120) and the last two identifying a narrow scope assignment (Architectural Cast in Place Concrete Formwork: 03120).
Since its introduction in 1963, MasterFormat has been widely accepted and promoted as an industry standard in the United States and Canada. It was first published as part of the ACSI Format for Construction Specifications,@ which evolved as the basis for the kUniform System for Construction Specifications, Data Filing, and Cost Accounting -- Title One Buildings@ published in 1966. Master Format has held the broadest possible acceptance as the filing system for materials and products. Its shortcomings as a method to describe a whole building are obvious to anyone who has had to review a contractor's Application for Payment using a Schedule of Values derived from the Specifications' Table of Contents. Early Division work is artfully front loaded with the hope that the minimal amounts allocated for end-of-project site work, concrete or masonry will be overlooked during the project's honeymoon phase billing. Classification errors in estimating, procurement, and coordination persist due to selective interpretation of broad scope requirements: Are lintels provided by the mason or furnished under Miscellaneous Metals? Access doors are required by mechanical trades, installed by as many as four different finish trades, and furnished under Division 08 (doors). Experience and local custom have largely compensated for many similar tactical difficulties.
ASTM E-2166 bridges these two systems by defining one additional level, the Type. A type is a kind of user-defined assembly that possesses a unique combination of function and components consistent with, and subordinate to, elements within the third level of the UNIFORMAT II outline. Elements that superficially appear to be similar are constructed with purposeful physical variations in order to accommodate a variety of ftmctional or situational requirements. For instance, some buildings may have fagade elements that respond to a base, shaft and crown design. Others K a y vary by floor level and compass orientation. Some may have an apparently uniform exterior while the backup construction varies wildly to conceal or protect interior activities.
The use of types is significant for several reasons. First, UNIFORMAT II does not inherently possess a method to differentiate among the many potential combinations of materials that might be assembled to describe satisfactorily the significant sub-set functions of an element. It would be quite unreasonable to expect to contrive a
comprehensive taxonomy of types given the staggering combinations of historic, stylistic, regional, and economic variations that occur in building construction. By setting this level to record the actual functional assembly, an accurate record of the building is maintained. The third level then remains a summary level, enabling comparisons across differing construction techniques or uses. Second, because the type is a sum of Master Format parts, the ability to step into a component level analysis is expedited to the shortest possible path. Third, the application of a type can contribute to support an overall strategy for faqade maintenance and component replacement. The capacity to evaluate elements types and components rapidly is useful to identify likely failure points or "w/eak links" in a given assembly.
CHADWICK AND MCJUNKIN ON FẴADE MAINTENANCE 1 13
When new installations have successfully survived the one-year statutory warranty, one can reasonably expect them to be reliable for the duration of the guarantee period.
With the guarantee period complete, we look to what experience would determine to be the probable maximum life, or the period of time when a minimum amount of upkeep is required to sustain the desired performance. As the component or system continues to age, the upkeep and maintenance needs continue in theory, to increase. An economic analysis can easily determine the cost-benefit o f repair-and-retaining, replacing, or use- in-place-until-failure at the end of its maximum possible life. Each choice presents risk.
Risk is mitigated with information. The difficulty arises out of selecting an appropriate depth of focus, or establishing a balanced view to state the problem for analysis. With this data structure, a facade can be separated into elements, types and parts to ferret out the instigating problem points. Conscientious application of General Conditions and General Requirements costs will contribute to the body of information from which a decision tree and action plan can be constructed. A reasonably knowledgeable building manager with appropriate engineering advice could implement a scheduled inspection cycle to optimize professional time on site to monitor conditions as well as make a convincing argument to fired a facade account with predictable draw-down events.
The other contribution made by the Standard Practice for Organizing and Managing Building Data is the consideration of joints or the connections between elements as their own typological set. As an outline comprised o f functional relationships, UNIFORMAT II tends to focus on each system as a discrete concept. Ultimately, to be functional, the elements must be connected to each other in some way to act as a coherent building.
These connections or joints are designed to maintain the integrity of the system by mitigating certain conditions within designed limits. The function of the joint is necessarily more complex than the types being joined. The notion e r a "joint" in every elemental system is tacitly presumed to reside in one or the other system and never dealt with directly. An awareness of the joint as a named entity helps to focus attention on its functional criteria. For instance, i f a basic function o f an exterior wall type is to keep weather out of the building, the joints must additionally accommodate movement, possibly provide galvanic isolation, and present an appearance consistent with an overall architectural vocabulary. The materials used to make joints are frequently unique to the joint and different than the materials comprising the basic types being joined. The useful
life and maintenance cycles of many kinds of joints vary sufficiently from the adjacent assemblies to merit scheduled attention.
The Standard Practice for OrganiZing and Managing Building Data was written with no explicit recommendation for a particular management mechanism. Based on the complexity of the building and the needs of the owner, physical tiles could be effectively managed under its data schema. Similarly, object-based programming and contemporary databases could be used to manage most buildings on a desk-top personal computer. The unprecedented power of inter-net and intra-net applications, however, provide a degree of resource management and sharing never before thought possible. Shadowing the
development of the Standard was the increasing popularity and evolutional improvements of eXtensible Markup Language, or XML.
XML evolved as a subset of Standard Generalized Markup Language (SGML), which is widely used in Europe in the publishing industry to assist in the electronic delivery and publication of text-based documents. As early as 1996 it was clear that SGML was too
114 BUILDING FẴADE, MAINTENANCE, REPAIR, AND INSPECTION
complex to be handled on the fly by web-based applications. Hypertext Markup Language (HTML), an application of SGML was too limited to handle digital presentations and could only present "images" of numeric tables - that is, the numeric data as-presented could not be manipulated with mathematical operations. XML is becoming the de facto Internet standard for representation of content optimized for Web delivery. It is a meta language for defining an unlimited number of specific markup languages, each of which may contain an unlimited number of tags, hence extensible.
Documents or records encoded to conform with XML have both a logical and physical structure. Logically, they consist of a hierarchy of named elements, which may be likened to fields, with nested elements akin to subfields. Each instance of a document has a single root element to which other elements are subordinate. XML provides for unambiguous identification of complex data structures that can be treated as objects. This is
accomplished through the Document Type Definition, or DTD. The DTD declares each of the permitted entities, elements and attributes and the relationships among them, forming a template for the logical structure of associated XML documents. It expresses the hierarchy and granularity of data, allowable attribute values and whether elements are optional, repeatable, etc. DTDs have been used to define a Biosequence Markup
Language, as well as ones for astronomy, chemistry, and mathematics. The Standard Practice for Building Data Management could reasonably form the core schema of the DTD for building data.
Using ASTM E-2166 as a style guide, building surveys and consultant reports could be produced in an easily readable organization that could be readily tagged for absorption into a larger data system. Each tagged data element is retrievable at the hierarchical level to which it was set, for "horizontal" comparison as one would expect in a conventional database. The set of attributes that can be associated with the XML data element however, give it an object quality, permitting it to be sorted according to any of the assigned attributes to yield a very fine grained analysis.
This application supports higher level documentation as well as detail analysis, andis intended to accept information on an "as available" basis. This is useful when a large portfolio of properties need to be initially evaluated or when invasive exploration can be deferred. An owner can specify observations to a degree of detail consistent with a directed level of interest. For instance, a binocular inspection with videographic documentation of a set of facades would provide a reasonable amount of baseline data to assign priorities and determine a plan of action. If conditions deserving urgent attention become apparent, they can be noted for subsequent investigation. As additional work is performed and further investigations made, the inventory of activity and degree of detail will eventually populate the component level. This higher level data collection also has obvious application in support of authorities-having-jurisdiction directing many professionals performing triage evaluations after natural disasters.
As the building stock inevitably ages and economic priorities shift, it becomes increasingly important for building owners to possess a lucid understanding of the conditions and costs associated with their buildings. The t e r m " stewardship" found in so many annual reports suggests a level of comprehension often promised but rarely demonstrated. In the face of construction techniques, legal requirements, and business priorities which become more complex daily, it will be the ability to understand and use the data at-hand that will dictate outcomes. Solid data with crisp organization will