4 High Risk and Complex Buildings
4.1.1 Risk means different things to different people. There are numerous definitions and interpretations of risk. Risk can be qualitative or quantitative. From an engineering perspective, risk is often represented as a function of the likelihood (probability) that a particular consequence (unwanted outcome) will occur. When using ‘risk’ in a guidance document, it is imperative that its use, interpretation and application are clear. Unfortunately, this is not the case with the Technical Handbooks.
4.1.2 The term ‘risk’ is used in a widely varying manner throughout the Technical Handbooks, in particular non-domestic, with differences between regulated areas (e.g., Structure and Fire), and within a single area (e.g., Fire). For example:
220.127.116.11 Section 1: Structure has four building ‘risk groups’: 1, 2A, 2B and 3, which relate to occupancy level, use, the number of storeys and floor areas.
18.104.22.168 In Section 2: Fire Introduction, the term ‘risk’ is used 22 times. Some examples include:
- “… where people may be asleep or where there is a particularly high risk.”
- “Occupants in buildings do not normally perceive themselves to be at risk from fire and are not usually aware of the speed that fire can spread.”
- “Protected routes of escape - throughout the document there are references to protected routes of escape these include: …places of special fire risk, …”
- “Certain types of buildings pose particular risks and require particular solutions. Additional guidance for three specific building types are grouped in three annexes; residential care buildings in annex 2.A; hospitals in annex 2.B and enclosed shopping centres in annex 2.C.”
- “Persons with obligations under Part 3 of the Fire (Scotland) Act 2005, as amended are required to carry out a fire safety risk assessment which may require additional fire safety precautions to reduce the risk to life in case of fire.”
- “Construction products are expressed as non-combustible low, medium, high or very high risk and explained in annex 2.E.”
22.214.171.124 In these examples there is, among others, reference to ‘particularly high risk,’ ‘special fire risk,’ ‘particular risks,’ ‘risk to life in case of fire’ and ‘low, medium, high or very high risk’ – each of which has very different meanings.
126.96.36.199.1 The reference to ‘particularly high risk’ implies life safety risk to occupants due to activity or vulnerability, such as sleeping.
188.8.131.52.2 The ‘particular risks’ refer to specific building types: residential care buildings, hospitals and enclosed shopping centres. This could be related to activity or vulnerability, but also to total number of occupants.
184.108.40.206.3 The ‘particular risks’ might also be the ‘places of special fire risk’ but that is unclear, since industrial or similar occupancies might present higher risk of fire occurrence, or higher risk of losses due to fire, but not necessarily higher risk to life from fire. In Section 2.1.8 of the Technical Handbook, paint spraying is the only ‘place of special fire risk’ noted; however, in the definitions, Appendix A, other ‘risks’ are listed.
4.1.3 This widely ranging use of the term ‘risk’, within the Technical Handbook guidance, can create challenges in interpreting, applying and the guidance for verification purposes.
4.1.4 As a general observation, it would be extremely helpful from a usability perspective, to develop and implement a common approach to the use of risk, risk classifications, risk levels, and the like, throughout the Technical Handbook. While outside of the scope of this research project, this warrants future attention.
4.1.5 In discussion below, suggestions are provided for how to characterise risk (focused on fire) and how a more uniform approach to classification for fire risk might be represented.
4.1.6 Complexity, too, can vary in meaning and by perspective. By one definition, complexity is “the state or quality of being intricate or complicated,” where ‘complicated’ is defined as “consisting of many interconnecting parts or elements” (https://en.oxforddictionaries.com/definition/complexity).
4.1.7 While arguably all buildings consist of many interconnecting parts and elements, and are therefore complex, many approaches to reducing complexity have developed over time, including standardised approaches to space utilisation (e.g., ‘standard’ office configurations), building components and systems (e.g., door sizes, structural systems, etc.), and construction.
4.1.8 As with the term ‘risk’ discussed above, the terms ‘complex’ and ‘complexity’ are used throughout the Technical Handbooks. With respect to fire, the most common applications seem to be as related to the following:
- Complexity of the building design, in part driven by the use, for example shopping centres, transportation hubs, multi-use buildings, and the like.
- Complexity of the fire engineering design, including use of multiple types of fire mitigation systems and strategies.
- Complexity as associated with the understanding and prediction of human behaviour in fire.
- Complexity of the analyses and tools of analysis.
4.1.9 These uses of the terms complex and complexity are explored in more detail below.
4.2 Characterising ‘High Risk’ Buildings With Respect to Fire
4.2.1 There has been much published in the literature regarding characterising risks in and of buildings from a wide range of hazards, including by the author, with a particular focus on fire (e.g., Meacham, 2004; 2007, 2010; Meacham and van Straalen, 2017). The term ‘higher risk residential buildings’ was recently introduced in the report by Dame Judith Hackitt (2018).
4.2.2 While a Scotland-specific risk characterisation process is ultimately needed, it is suggested that to begin with consideration of fundamental components that have been identified in previous efforts: hazard factors, risk factors and importance factors be undertaken.
220.127.116.11 Hazard factors are developed in response to such questions for example, what is posing the risk, what is the nature of the harm, where is the hazard experience, where and how do hazards overlap? Given such considerations, a set of hazard factors for buildings can be developed, such as:
- The nature of the hazard
- Whether the hazard is likely to originate internal or external to the structure, and
- How the hazard may impact the occupants, the structure, and/or the contents.
18.104.22.168 Risk factors are developed in response to such questions as who is exposed, which groups are exposed (i.e., all of the population, sensitive populations, etc.), what characteristics present the risk, what qualities of the hazard might affect judgments about the risk? Given such considerations, a set of risk factors for buildings can be developed, such as:
- The number of persons normally occupying, visiting, employed in, or otherwise using the building, structure, or portion of the building or structure.
- The length of time the building is normally occupied by people.
- Whether people normally sleep in the building.
- Whether the building occupants and other users are expected to be familiar with the building layout and means of egress.
- Whether a significant percentage of the building occupants are, or are expected to be, members of vulnerable population groups.
- Whether the building occupants and other users have familial or dependent relationships.
22.214.171.124 Importance factors relate to the real or perceived importance of a building to a community, i.e., what are key reasons as to why a community may deem a building, or class of buildings, to be important to community welfare perspective. Key importance factors include:
- The service the building provides (e.g., a safety function, such as a police or fire station, or a hospital)
- The service the building provides in an emergency (e.g., an emergency shelter, hospital, communications facility, or power generating station)
- The building’s social importance (e.g., a historic structure, a church or meeting place), or
- The hazard(s) or risk(s) the building poses to the community, not just its occupants (e.g., chemical manufacturing facilities or nuclear power generating facilities).
126.96.36.199 By taking such an approach, one can develop ‘risk groups’ that define the major considerations by which buildings in a jurisdiction might be considered. This in turn can lead to more uniform ‘risk mitigation’ requirements (or recommendations / guidance) to be applied across the building stock. An example of using ‘risk groups’ is shown in Table 4.1, which is excerpted from the International Building Code (IBC) in the USA, as based on the structural design code, ASCE 7.
188.8.131.52 This same fundamental structure is incorporated into the International Code Council’s ICC Performance Code for Buildings and Facilities (ICC Performance Code) as well, although the ICC Performance Code is not widely adopted in the US.
4.2.3 There are other approaches used in various countries as well. In some cases, specific Occupancy Groups (Use Groups, Building Classes, etc.), such as Assembly, Business, Hazardous, Healthcare, Industrial, Institutional, Mercantile, Residential, etc. These approaches typically include implicit characterization of risk, but not explicit (e.g., ‘healthcare’ occupancies might have more fire protection measures, since the occupants are viewed as more at risk / vulnerable). However, this approach can lead to numerous sub-categories, as well as special consideration and/or exceptions.
184.108.40.206 For example, in the International Building Code (IBC) in the US, there are 8 major Use and Occupancy classifications, some with as many as 5 sub-categories, and an additional set of special detailed requirements for 24 specific uses. In such a system, it can be quite difficult to assure that the understanding of implicit levels of risk / safety are fully understood.
220.127.116.11 However, the two approaches need not be mutually exclusive. It is possible to map the specific use classifications in the IBC to the Risk Groups in the IBC, if so desired, although this is arguably a redundant step.
4.2.4 A more quantified risk approach is being considered in some countries, such as Australia and the Netherlands.
18.104.22.168 In Australia, the approach being explored is built around quantifying the individual and societal risk to life from all sources, quantifying the individual and societal risk to life from hazards that impact buildings (i.e., for which mitigation via building regulation / building design are intended to address), and establishing benchmark levels of tolerable risk that a building design should meet.
22.214.171.124 In the Netherlands, the approach being explored considers the probability of life loss in a building, given a hazard event or system failure (e.g., structural system failure). The approach is modelled on the risk-informed approach in the Eurocodes for Structure.
126.96.36.199 In both cases, there is a significant reliance on the data used for benchmarking and the methods used for analysis.
Table 4.1 Example of Risk Group Approach (IBC, 2012)
Nature of Occupancy
Buildings and other structures that represent a low hazard to human life in the event of failure, including but not limited to:
Buildings and other structures except those listed in Risk Categories I, III and IV
Buildings and other structures that represent a substantial hazard to human life in the event of failure, including but not limited to:
Exceed maximum allowable quantities per control area as given in Table 307.1(1) or 307.1(2) or per outdoor control area in accordance with the International Fire Code; and
Are sufficient to pose a threat to the public if releasedb.
Buildings and other structures designated as essential facilities, including but not limited to:
a. For purposes of occupant load calculation, occupancies required by Table 1004.1.2 to use gross floor area calculations shall be permitted to use net floor areas to determine the total occupant load.
b. Where approved by the building official, the classification of buildings and other structures as Risk Category III or IV based on their quantities of toxic, highly toxic or explosive materials is permitted to be reduced to Risk Category II, provided it can be demonstrated by a hazard assessment in accordance with Section 1.5.3 of ASCE 7 that a release of the toxic, highly toxic or explosive materials is not sufficient to pose a threat to the public.
4.2.5 A similar approach to that described above in the ICC is also used within the Eurocodes for Structure, but in this case, the approach is reflected as consequence classes.
188.8.131.52 The following table presents the definitions of the three consequence classes from the Eurocodes (EN 1990, Table B1).
Table 4.2 Eurocode Consequence Classes (EN 1990, Table B1)
High consequence for loss of human life, or economic, social or environmental consequences very great
Grandstands, bridges, public buildings where consequences of failure are high (e.g., concert hall)
Medium consequence for loss of human life, or economic, social or environmental consequences considerable
Residential and office buildings, public buildings where consequences of failure are medium (e.g., office building)
Low consequence for loss of human life, or economic, social or environmental consequences small or negligible
Agricultural buildings where people do not normally enter (e.g., for storage), greenhouses
184.108.40.206 It is observed that the most significant differences between the ICC and Eurocode approach are that the ICC approach breaks down the specific ‘risk factors’ in more detail (e.g., number of people at risk), while the Eurocode approach focuses more on generalised consequences (e.g., consequence is high). A combination of approaches might be worth considering.
4.2.6 As Australia explores a more quantified approach to risk as a basis of building performance, it has been suggested that an approach that identifies specific risk factors of concern be considered. The following reflects some thinking in this regard.
220.127.116.11 As outlined above, an independent review of the NSW Building Professionals Act 2005 was recently undertaken and the findings were published in a 2015 report (http://bpb.nsw.gov.au/sites/default/files/public/Attachment%20A%20-%20Final%20Report.pdf, last accessed 12 April 2018). This review identified many issues which have parallels to the Scottish situation. The following is excerpted from the report (underline and highlight added by the author):
18.104.22.168.1 In the section on proposed fire safety reforms, the following is noted (pp240-241):
“The key to reform in this area is to ensure that properly qualified and experienced persons are accredited to design, install, commission and maintain fire safety systems with particular attention given to alternative fire safety systems. This will need to be supported by a risk based review process to evaluate complex or significant fire safety systems at the design stage. This will need to draw on suitable independent parties which could include the following:
- Independent review by an accredited fire safety engineer or a company involved in fire safety engineering
- Independent Peer Review Panel which would be a group of accredited certifiers including a fire engineer and an accredited person with fire-fighting experience
Criteria will need to be developed for the circumstances that would trigger an independent review and for when each of the three review mechanisms would be employed. The Society of Fire Engineers has provisionally suggested that fire safety engineering relating to the following circumstances should be considered for an independent review:
- Large infrastructure projects
- Buildings over 25 m in effective height
- Assembly buildings containing more than 1000 occupants
- Buildings containing an atrium which connects more than three stories
- Buildings where the main structure is of exposed steel or timber in lieu of a designated fire resistance level
22.214.171.124.2 It is observed that the suggested criteria are interesting in that they represent the type of criteria that could be considered in Scotland as triggering the need for review by a Chartered Fire Engineer, or even, as suggested for NSW, by an independent review panel.
4.2.7 Ultimately, any approach to defining ‘high risk’ building in Scotland should begin with a discussion on defining and characterizing risk, and then moving on to categorizing or quantifying risk, as befits the selected model.
4.2.8 Consideration of existing classification(s) of risk in the Scottish system would be a likely basis of such an effort (e.g., looking to ‘places of special risk’ and buildings that ‘pose a particular risk’ as discussed in the Technical Handbooks).
4.3 Characterising Complexity for Fire Safety Design
4.3.1 Raman suggests that in modern buildings, complexity comes from four primary sources (available for download from http://src.holcimfoundation.org/dnl/32e9279c-84dd-4f2e-9547-db6d983bf3f9/F10_BlueWorkshop_Paper_RahmanMahadev.pdf, last accessed 12 April 2018):
- Sophisticated building components
- Sophisticated systems
- Multi-disciplinary integration
- The desire for endless novelty in the built form
4.3.2 Raman suggests that “buildings like this represent, perhaps, no more than 10% of all the building activity that takes place at any given time, but they create much of the excitement in architectural circles. They also offer the greatest challenges in terms of managing complexity and risk.”
4.3.3 It is suggested that this taxonomy represents a good starting point, particularly for new construction. There are also considerations associated with the complexity of tools and methods used for analysis and design of systems and performance.
- The sophistication of methods of analysis (in particular, computational tools, such as computational fluid dynamics (CFD), finite element (FE) software, and computational evacuation software.
- The integration (or not) of the various software tools in adequately assessing the holistic performance of a building and its systems.
4.3.4 Added to this might be issues associated with existing construction, including the following:
- Integration of new construction into existing built environment (in particular within dense urban environments)
- Sophisticated ownership or tenancy issues associated with the integration of new construction into existing, including boundaries, pedestrian flows between spaces, and user responsibilities (e.g., systems / space maintenance)
4.3.5 There are also attributes of the design and procurement processes that introduce complexity into the building design and verification process.
- Systems in which there is a not single, clearly defined ‘responsible’ entity for the design, which assures that the building and its systems are appropriately integrated and implemented in the final operational building.
- Systems in which there is no requirement by designers / engineers to assure that the ‘as-built’ building and its systems meet the design strategies and associated requirements.
- Systems in which there are few requirements for inspections, testing and commissioning of systems, and other such measures to control quality during construction.
- Systems in which ongoing maintenance and proper operation of the building and its systems are not routinely audited for compliance with the design strategy.
4.3.6 In considering the complexity of systems (including buildings, which are complex systems of systems), and the associated reliability of the systems in delivering the expected performance when needed, the extent of interrelationships and dependencies is important.
4.3.7 In looking at risk associated with complex systems, Perrow (1984) pointed out that complex, tightly coupled systems have more risk of failure than loosely coupled systems. This is largely because of the high level of reliance that each component in the system will delivers its expected function when needed, and if one component fails, the whole of the system is more likely to fail. Unfortunately, attempts to improve safety and reliability through more effective regulation introduces further complexity, intensifying non-linearity and increasing risks, although different than the initial risk challenge (Burns and Machado, 2010).
4.3.8 Arguably, building regulatory systems are themselves complex, socio-technical systems (Meacham and van Straalen, 2017). In order to manage complexity in the system, a number of factors are important to have, including:
- A well-defined building regulatory system, including the interconnections between actors, institutions and technologies.
- A well-defined set of verifiable building performance expectations.
- Education, training, and associated resources to facilitate the required levels of qualifications and competencies of the actors to deliver on design, regulatory, operational and related needs.
- An appropriate set of design, regulatory, and operational data, tools and methods to facilitate delivery of the expected building performance.
4.3.9 These concepts align well with those of Schalcher, who suggests that in order to manage complexity in planning, design and construction, the following maxims apply (available for download at https://src.lafargeholcim-foundation.org/dnl/901ed18a-96ca-4904-a23c-c2620281c611/F10_BlueWorkshop_Paper_SchalcherHansRudolf.pdf, last accessed 16 April 2018):
- Consider a building as a strongly interrelated element of a preceding, extremely complex, human and natural system.
- Plan and design not only for the initially defined use but also for unexpected transformations.
- Foster diversity of use, layout, materials and technologies.
- Apply the principle of integration instead of deconstruction and segregation.
- Achieve economy of means and reduce metabolism by multiplicity and multifunctionality (i.e., one item fulfils more than one purpose).
4.3.10 If we consider the process of planning, designing and executing a building we should be prepared to:
- Take decisions on the basis of fuzzy, i.e., incomplete and uncertain, information.
- Involve internal and external stakeholders at an early stage.
- Manage projects through leadership, team work and forward coupling.
4.3.11 Each of the above seem sets of perspectives is pertinent to the issue of dealing with complexity of buildings / fire engineered designs in buildings in Scotland.