©) Taylor Wood Material Science & Engineering ISSN: 1748-0272 (Print) 1748-0280 (Online) Journal homepage: [Link]/journals/swo020 Structural robustness and timber buildings - a review Johannes A. J. Huber, Mats Ekevad, Ulf Arne Girhammar & Sven Berg To cite this article: Johannes A. J. Huber, Mats Ekevad, Ulf Arne Girhammar & Sven Berg (2019) Structural robustness and timber buildings - a review, Wood Material Science & Engineering, 14:2, 107-128, DOI: 10.1080/17480272,2018.1446052 To link to this article: [Link] 8.1446052 © 2018 The Author(s), Published by Informa UK Limited, trading as Taylor & Francis Group PR) rubles one: 13 Mar 2018 CF submityour article to this journal CZ ltl Article views: 8493 LEG view relate ariies @® view crossmark data CB) citing artices: 24 View citing articles Full Terms & Conditions of access and use can be found at ttps:/[Link]/action/journalinformation?journalCode=swoo20(000 MATERIAL SCENCE & ENGNEERING 2018, VOL. 14, NO. 2, 107-128 htsvidl erga r0s1 76502722018 16082 Taylor & Francis REVIEW ARTICLE BOPEN ACCESS ® os Structural robustness and timber buildings - a review Johannes A.J. Huber ®, Mats Ekevad ©, Ulf Ane Girhammar ® and Sven Berg © Division of Wood Science and Engineering, Department of Engineering Sciences anc Mathematics, Luled University of Technology, Skellfte, Sweden STRACT Timber buildings are Increasing in their dimensions. Structural robustness is imperative for all buildings and specifically important for tall buildings. Lives can be saved if disproportionate collapse can be avoided after a catastrophic event (e.g, accident, terrorism). The literature about robustness is comprehensive concerning concrete and steel buildings, but is rather limited regarding timber. This paper reviews robustness in general and robustness of timber buildings in particular. Robustness is an intrinsic structural property, enhancing global tolerance to local failures, regardless of the cause. A deterministic approach to assess robustness is to remove certain load- bearing elements from the structure and compare the consequences to given limits, Design methods for robustness may be direct by assessing effects of loca falure, or indirect by following guidelines. For robust timber buildings, the connections are the key aspects. Usually, metal connectors may provide the required joint ductliy. For robust light timber-frame construction, rim beams may be designed. For timber posts and beams and cross laminated timber, guidance regarding robustness is scarce, but in some aspects they seem to be similar to steel frames and precast concrete. Future research should assess the capacity of connections, and evaluate the adequacy of seismic connectors for robust timber buildings ARTICLE HISTORY Received 21 December 2017 Revsed 16 February 2018 ‘Accepted 28 February 2018 Kerworos Robustness, tibet, dlsproperionateealpse progressive calpse, Ekernatve fad pat, famage tolerance 1. Introduction Building structures are expected to protect human lives and provide shelter throughout thei lifetime. To full their protec- tive function, buildings are required to survive both expected and unexpected load scenarios, Established design rules and codes such as the Eurocodes aim to ensure that buildings withstand expected loads during the building's lifetime. Treat- ing unexpected loads requires a different approach as neither the magnitude nor the location of these loads is known. Robustness deals with questions of how a building structure may tolerate unforeseen load events resulting in local failure and still fulfil ts function to protect human lives. The more severe the expected consequences of a collapse of a building are, eg. the more people who may be harmed, the more acute is the matter of robustness, The steel and concrete bulld- ing community has confronted this challenge for a longer time than that for timber buildings. Therefore, more extensive research into robustness has been conducted regarding steel and concrete and comprehensive literature is availabe. Historically, builders lacked exact analytical models, but in many cases they could produce sufficiently robust structures to endure a long lifetime (Knoll and Vogel 2008). In modern times, refined analytical modelling capability has emerged, bullding processes have become more complex and struc- tures more optimised. Some modern buildings have failed in spite of the available tools because some factors have not been considered in the models (Knoll and Vogel 2009) Robustness may not be implied in a modern building without explicit checks (ISE 2010). ‘A well-known event which initiated research in the field of structural robustness is the case of Ronan Point in London, UK in 1968. There, a gas explosion in a corner apartment of a 22 storey concrete building led to the collapse of the entire building corner (ISE 2010; Byfield et al 2014). Another charac- teristic event is the collapse of a large part of the Murrah Federal Building in 1995 in Oklahoma City initiated by a car bomb (Byfield et al. 2014). The World Trade Center collapse in New York after a terrorist attack in 2001 provided the latest impulse to this research field (ISE 2010; Kokot and Solomos 2012) ‘The more storeys a timber building has, the higher is its consequence class according to Eurocode 1 (2006). Timber buildings of higher consequence classes are confronted with the challenge of structural robustness in order to provide safe living spaces, The literature in this respect regard- ing timber is stil rather limited. This paper provides a review of structural robustness in general and its implementations in timber structures in particular 2. Materials and method A literature study has been performed, The search for ltera- ture was mainly performed using online search portals linked to electronic sources of different publishers. One CONTACT Johannes AJ. Huber © johannes uber se @ Division of Wood Scence and Engiesing, Depoiment of Engineering Sciences and Mathematics, ale Universy of Technlogy, S937 47 Shelled, Sweden © 2018 The Autre Pulse by forma UK Ute, wag 2 ay & rads ious108 © LALMVRERETAL focus was put on finding material describing robustness in general such as other reviews. An effort was also made to find a sufficient quantity of sources with timber-specific considerations. The terminology used for structural robustness was briefly revised to develop a clear understanding of the Underlying ideas. Different concepts of analysis and quant fication were then examined. In the next step, general design strategies independent of the material used were reviewed. Finally, analyses and design recommendations regarding structural robustness specifically for timber build- ings were reviewed. 3. Defi ions ‘The terminology used in the literature in the context of struc tural robustness of buildings seems to be inconsistent (Agarwal and England 2008; Arup 2011; Starossek and Haber- land 2011; Brett and Lu 2013). In the following, certain estab- lished terms are briefly described and suitable definitions are followed throughout the text. In this paper, damage and failure are used to refer to both structural components and the whole building, Starossek and Haberland (2010) define damage as a deviation from the design state which may lead to partial loss of functionality, and failure as the total loss of functionality. Partial loss indi- cates that some reserve functionality is still available. If 2 buliding is concerned, damage refers to the functionality of the entire building. In this paper, a collapse is regarded as the failure of a substantial part of a building. Disproportionate collapse and progressive collapse are fre- quently used terms in the context of robustness. A dispropor- tionate collapse is a structural collapse where the initial cause and its subsequent extent stand in a disproportionate relationship to each other (Starossek and Haberland 2010; Arup 2011). A progressive collapse is a structural collapse, beginning with an initial component failure, leading to sub- sequent failure of components that were unaffected by the Initial failure (Starossek and Haberland 2010). Agarwal and England (2008) say that the initial failure spreads like @ chain reaction in such types of collapse. They also say that 2 progressive collapse may be disproportionate in its extent but the opposite may not necessarily be true, The term pro- agressive describes how a collapse develops and the term dis- proportionate describes how much damage the collapse leads to compared to the initial damage. Unfortunately, the two terms are often used interchangeably in the literature, In this work, the terms are used separately according to the above definitions. The term robustness is used ambiguously in different pub- lications, although the structural concept behind the word seems to be consistent in the literature (Brett and Lu 2013). Starossek and Haberland (2010) propose clear distinc- tions and definitions to avoid ambiguity. To distinguish robustness from other properties, they present a qualitative ‘model based on an abnormal event acting on a structue, E, Which may lead to an initial local damage, D, which may in turn provoke a disproportionate failure, C. They then decom- pose the probability of a disproportionate collapse, P(O), as a result of an abnormal event, into the following probabil- ities: PCC) = CID)» PIE) » PE) a where P(E) Is the probability of occurrence of an abnormal event, P{D|E) is the conditional probability of initial damage given the abnormal event and (CID) is the con- ditional probability of disproportionate spreading of failure given the initial damage Equation (1) was initially introduced by Ellingwood and Dusenberry (2005) and seems to be established in the litera- ture (Kokot and Solomos 2012; Brett and Lu 2013). Starossek and Haberland (2010) refer to P(E) as exposure, to P(D |E) as vulnerability and to P{CID) as robustness, which together make up a building's collapse resistance. They explain that only vulnerability and robustness are structural properties which may be affected by structural engineering. Whether for not damage is spread is described as being an intrinsic property of the given structure and therefore independent of any abnormal event. Examples of useful definitions for robustness are the following: Insenstivty ofa structure to inal damage. A structure Is bust i aninital damage doesnot lad to elsproporionate collapse. (Sta cossek and Habetland 2010) the terms structural robustness or robustness are used to describe a quay ina structure of insenitvty to local fire, in which modest damage (whether due to accidental or malicious action) causes only a simlarly medest change In the structural behaviour. up 2011), ‘Aualty in a strueture/stuctura system that desebes its ably to accept 2 certain amount of damage without that structure fling to any great degree, Robustness implies insensitivity to local fare. (SE 2010) ‘The definition in the Eurocode is more event-specific: the ability ofa structure to withstand events like ie, explosions, Impact oF the consequences of human etter, without being damaged 10 an excent disproportionate to the orginal cause (Eurocode 1 2006) Concordantly, robustness may be defined as an intrinsic property of the structure alone which enhances a building's global tolerance to local failure, independent of the failure ‘cause. This definition is followed throughout this text. Synonyms for robustness found in the literature may be progressive collapse resistance, resiience or insensitivity to local [Link] suggested that such synonymous expressions be avoided to avoid ambigu- ity. In some contexts, the wording structural robustness instead of robustness alone may be more appropriate. 4. Analysis and quantification In general, the analysis and quantification of robustness may be based on (1) a risk analysis, (2) a reliability analysis o (3) a deterministic analysis (Kirkegaard et a. 2010; Ciamar et a 2011b; Sorensen 2011; Brett and Lu 2013; Kovecsi 2014;Chen et al, 2016). Risk and reliability analyses are probabilistic approaches and thus take into account probability distr- butions regarding building exposure or material parameters AA deterministic analysis may be conducted in a pragmatic manner and is needed as a complement to a probabilistic analysis (Starossek 2006). Both probabilistic and deterministic analyses may yield measures to quantify robustness. The generic formulations of robustness measures are similar and are based on the insensitivity of the system to a disturbance in 2 variable (Brett and Lu 2013). An overview of measures and a proposal for a comprehensive measure are given by Brett and Lu (2013). The probabilistic approaches are described briefly in the following subsections, but this paper focuses on deterministic considerations and the deterministic approach is therefore described in greater detail 4.1, Risk analysis Baker et al (2008) propose that the robustness be assessed by ‘modelling the total risk. Their reasoning is based on paths in decision trees like that shown in Figure 1. The path starts at the point in the left ofthe figure, and one of many possible exposures before damage (path Eso in the figure) is assumed. If no damage occurs (path in the figure) as a result of the exposute, the total risk for this path is zero. if damage occurs, one of several possible damage states mate- rialises (path D inthe tree) Baker et al (2008) explain that for each of the damage states, a system failure (path F) may or ray not result (path F) with a certain probability. On the right end ofthe tree in Figure 1, consequences are associated with the vatious paths. A damaged state always contains direct consequences (Cp. in the figure) which are directly related to the initial damage. Indirect consequences (Cra in the figure) are additional consequences of the intial damage and may only arise if system failure occurs. Baker eta. (2008) say thatthe consequences Cov and Ca may be quantified by deaths, injuries, inconvenience for the users or financial costs. Cor and Cg ate then used to quantify the risks due to direct and indirect consequences respectively as total probabilities along all possible paths. The relationship between indivect and direct risks is then used to yield 2 robustness index. 4.2. Reliability analysis: The reliability of a structural system quantifies the system's probability of performance over a building's service life (Kahler 2006). The probability of performance is the comp- lement to the probability of failure. Kohler (2006) says that, ina reliability analysis, fallure (Le. norperformance) is Figure 1. Decson te, adapted fom Baker el 2008) {WOOD MATERIAL SCIENCE & ENGINES we © 09 defined by any of the limit state functions reaching a value smaller or equal to zero, A limit state function may be based on a chosen serviceability, damage or ultimate limit state of the structure. Kohler (2006) states that the commonly used reliability index, denoted as 8, is linked to the probability Of failure, denoted as P), via the cumulative standard normal distribution function, denoted ©: Q where B = 0 for Py = 0.5 and increases for lower values of Py For P; > 0,8 > eo and for P; -> 1, B > —e0, The choice of a suitable target reliability index for a given situation is described for example in the report of the Joint Committee fon Structural Safety (2001). Based on the relationship between the reliability index of an intact (undamaged) system, denoted Byaqy and the reliability index of the same system in a damaged state, denoted Basmagees Frangopol and Curley (1987) introduced a redundancy index for damaged structural systems denoted Bx. Their definition of damage includes any strength deficiency in the structure. They further define redundancy as a state where the failure of no single component will lead to a collapse of the structure. Frangopol and Curley (1987) define Bi as follows: ~ Bron: ~ Beans ° Br IF complete damage is assumed for the damaged state, then Beanages =~ and thus By = 0. Ifthe structure's reliability is unaffected by the assumed damage, then Beanased = Prine 290 thus By =e. ‘The redundancy index Bis also used by certain authors to indicate the level of robustness (Kitkegaard and Sorensen 2008; Ciemar et al, 2011b; Sorensen 2011). Some papers suggest that a relabilty-based robustness index. denoted Frans may be writen as a direct relationship between the reliability indices for the intact and damaged systems (Cizmar et al. 20112; Sorensen 2011): @ This index thus assumes the value 1 if the reliability index of the damaged system is the same as that of the undamaged system. For a lower reliability of the damaged system, Bivragee —> —02 ad thUS laon p —* 22 4.3. Deterministic analysis In a deterministic analysis of robustness, the structural response of the building to the assumed initial damage or to ‘a specific exposure is evaluated. The term notional damage is used in the literature to indicate hypothetical damage to a structure. Notional local damage of the global structure may for example be the failure of a single stvuctural element in the building (e.g. the failure of a column). Specific exposures may for example include explosion, malicious (terrorist) attack, earthquake or fire. Analysis approaches which consider notional damage are referred to as scenariovindependent, and0 © LALMVBERETAL approaches which consider a specific exposure as scenario- dependent (Arup 2011), In a scenario-independent approach, the exposure which caused the initial damage is disregarded in the analysis in ‘order to evaluate the building's builtin ability to sustain damage of any type (Arup 2011). The sudden removal of @ load-bearing element is the most accepted method in a scen- ario-independent approach (Arup 2011). ISE (2010) indicate that the removal of a load-bearing element may refer to 2 column, wall or support structure for columns or walls. Elements should be removed one at a time, on each storey, unless it can be shown that element removal in different storeys leads to similar results. Ina scenario-dependent analysis, a specific exposure on a building is considered. This analysis may be used to demon- strate structural robustness in specific events, whereas the scenario-independent approach may be used to establish a baseline of robustness (Arup 2011). This paper focuses on scenario-independent approaches. 43.1. Alternative load path analysis To analyse the structural response of a building to the notional removal of an element, an alternative load path analysis (ALPA) may be performed (Ellingwood et al. 2007; Arup 2011; US. DoD 2016). The objective of an ALPA is to assess how loads are absorbed along altemative paths in the structure after the initial damage, and to quantify the extent of the collapse progression, 43.1.1 General concept. In a model of the building, a load- bearing element is notionally removed, and the structural consequences are studied. As the notional removal is arvied out in a sudden manner, dynamic load effects should be taken into account (Arup 2011). Debris loading from falling parts during collapse need to be taken into account in some cases (ISE 2010; Arup 2011). The extent of the removal is often specified by the nominal length, This length usually does not exceed 2.25H where H is the storey height, but for an external masonry, timber or steel stud wall, the length between vertical lateral supports (eg. columns or perpendicular walls) should be used (Eurocode 1 2006; ISE 2010; Arup 2011). The choice of removed elements ‘ean be summarised as follows (ISE 2010) + The removed element may be a support column, the nominal length of a load-bearing wall section, or structures supporting these two elements. + Elements should be removed one at a time, on each storey, Unless it can be shown that element removal on different storeys leads to similar results + If several columns are located within a diameter of nominal length, they should be removed simultaneously. + In comers, the length of load-bearing walls removed should be H in each direction, but not less than the dis- tance between expansion or control joints. ‘A reason why to use 2.25H as the maximum removal length for load-bearing walls was dificult to find in the litera- ture, Since this value represents an upper limit, it remains generally unclear how much wall length should be removed. Using the maximum would be conservative, but in some cases the removal of a length between vertical lateral supports could be sufficient, even for internal walls. ‘Arup (2011) argues that the design against the removal of a single column or a nominal length of a load-bearing wall results in sufficient robustness for most buildings, and that these uniform removals could standardise robustness. Detailed guidance regarding loads which should be applied in an ALPA can for example be found in the GSA (2013) and US. DoD (2016) codes, In these codes, the . H, Sorensen, J. D, Comat, D. and Dietsch, P. (2010) Robustness evaluation of timber steuctues ~ results ffor EU COST WOOD MATERIAL SCENCE KENGINEERNG @) 127 action E55:WG3, In PJ. da Sousa Cruz (ed) Structures & Architecture: ISA 2010 ~ 15 Intemational Conference on Structures & Architecture, 21-25 July 2010, Gurmaraes, Portugal (CRC Pres), pp. 129-131 Kirkegaard, P.H, Sorensen, J.D, Cizmar,D. and Raj, V. (2011) System relabilty of timber structures with ducile behaviour. Engineering Siractres,33(11), 3093-3098, Knoll F and Vogel, T. (2008) Design for Robustness (Zurich: Intemational Association fo Bridge and Strtual Engineering (ABSE) Kehler 1 (2006) Reliability of timber structures. PAD thesis. Swiss Federal Insttte of Technology. Kokot 5, and Solomo, 6. (2012) Progressive collapse rk analysis Literature ‘sure, relevant construction standards and guldelines (Luxembourg Publiatons Office of the European Union, EUR 25825 EN), Kovees T2014) Membrane ation of lbs in framed structures in case of ‘accidental column loss, MS thesis, Poitehnica University Timisoara Kuhlmann, U, Jaspat, JP, Vassar, ©, Weynand, K and Zandonin, R (2006) Robust structure by joint ductility. I ABSE Symposium Report Responding to Tomonew’s Challenges in Structural Enginerng, Budapest, Val. 92 (Zurich: Intemational Association for Bridge and Stuetual Engineering (ABSE), 9p. 1-8 Kuhlmann, U, ale, L. Jaspar, J-P, Demonceau, JF, Vassar Oy Weynand, k, Zier, C, Busse, €, Lendering, M, Zandoniny, R. and Baléassno,N. (2009) Robust Structures by Joie Ductility (Laxembourg Office for Oficial Publications ofthe European Communit). Kwasniewski L, aude B.A, Pereira, M, Bucur C and Glows M2008) ‘Modeling ane analysis. In J Kohler, Narastmhan and MH Faber ed) Proceedings ofthe Joint Workshop of COST Actions TUOSO? and ESS, 21-22 September 2009, udiana (Brucls: COST Ofcel, op. 91-102 Lamrence, A. (2014) Recommendations forthe design of complex indeter ‘inate timber structures, nS Aicher,H-W. Reinhardt and H, Garreent (eds) Materials and Jomts in Timber Structures (Dordrecht: Springer Netherlands), pp. 129-134 Lawson, P.M, Byfield, M. P. Popo-Ola, 5. 0. and Grubb, P. J. (2008) Robustness of light steel frames and modular construction Proceedings of the Insitution of Gil Engineers ~ Stuctres and usdings, 1610), 3-16, Lupoae, Mand Bucur, C2008) Use of applied element method 10 simulate the collapse of a building. In Annual Symposium of the Insite of Solid Mechanics SISOM 2009 and Sesion ofthe Commision of Acoustics, 28-29 May 2008, Bucharest (Bucharest: Romanian Academy), pp. 13-18, Malo, KA, Sir, Jan Elingsbe,P. (2011) Quantfying ducity inter structures. Engineering Souctures, 33(11), 2998-3006, -Marjarishll, 5. M2008) Progressive analysis procedure for progresive collapse. Journal of Perfomance of Constructed Facies, 18(2), 79-85. Marjanishul, 5. Mt and Agnew, , (2008) Comparison of various pro ‘cedures for progressive collapse analysis. Journal of Performance of Constructed Facites, 208), 365-374 pid Bt, H, Curtie, N. ane Tannert, T2016) Assessment of disproper: Lonate collapse for mult storey crosslaminated timber buildings In 2, Ebethardstemner, W. Winter, A. Fedai and M, Poll (eds) World Conference on Timber Engineering 2016, 22-25 August 2016, Vienna (Wienna: TU Verlag), pp. 3885-3701 pid ita, M, Curr N-and Tanner T. (2017) Reliability analysis and eis Proportionate collapse for mult-storey crossiaminated timber bul tng, In 16th World Conference on Earthquake Fngineering (VEWCEE 2017}, 9-13 January 2017, Santiago, Chile (Kanpur National Information Centre on Earthquake Engineering). Muneh-Andersen, J. and Dietseh, P. (2011) Robustness of largespan ‘umber roof structures ~ two examples, Engineering Structure, 33111), 33-3117, [Nimse, RB, Joshi, D. D. and Patel, PV. (2014) Behavior af wet precast bbeam column connections under progressive collapse scenario: An experimental study. International Joumal of Advanced Structural Engineering, (8, 148-158, Rajte, V, lamar, D. Kirkegaard, P. H. and Sorensen, J. D. (2010) Robustness analysis of tmber tus svucture, A. Ceccott (ed) Word Conference on Timber Engineering 2010, 20-24 June 2010, Riva el Garda, Haly (va del Gard, Italy: Trees and Timber hatte, National Researeh Council, pp. 1256-1263JAA HUBERETAL Sasani, M. (2008) Response of 2 reinforced concrete infiledframe struc ‘ure to removal of two adjacent columns, Engineering Structures, 30 (9), 2476-2081, Seim, W, Hurnmel J. and Vogt, . (2014) Earthquake design of timber structures ~ Remarks on force-based eesign procedures for diferent wall systems. Engineering Stuctures, 76, 124-137. Smith, JW. (2003) Energy approach to assessing corrosion damaged structures Proceedings ofthe Insbtuton of Cl Engineers ~ Structures ‘nd Buildings, 1562), 121-130. Smit Land Frangi, A (2008) Overview of design issues for tll timber bullings. structural Engineering Intemational, 182), 141-147 Smith, T, Watson, C, Moroder, D, Pampani, 5. and Buchanan, A (2016) Lateral performance of a Pres-Lam fame designed for gravity loads Engineering Stuctares, 122, 35-1 Sorensen, J.B. 2011) Framewore for robustness assessment of timber structures. Engineering Structures, 33(17), 3087-3092. Starossek U (2006) Progressive collapse of structures: nomenclature and procedures, Structural Engineering Interational, 162), 173-117 Starossek, U. (2007) Typology of progressive collapse, Engineering ‘Smuctures, 2919), 2802-2307. Starosse, U. and Haberland, M (2010) Disproportionate collapse: term nology and procedures. Jounal of Performance of Constructed Feces, 24(6, 519-528, Starossek, U. and Haberland, M. (2011) Approaches structural robustness, Stuctue and infasteactre Engineering, 70 e2s-631 measures of Steha, Lane Bores, K. (2004) The influence of nail ductlity onthe load capacty of a glum wuss stucture, Engineering Strctures, 2616, Tagel-Din,H. and Rahman, N. A. (2008) The applied element method. Sructure Magazine, Apri 30-33. Trelandesson, 5. and Hon, D. (2009) Behaviour and modeling of timber structures with reference to robustness. In. Kober, H, Narasimhan and M.A, Faber (eds) Proceedings of the Joint Workshop of COST Actions TWoeor and E55, 21-22 September 2008, Ljublana (Brussels: COST Office, pp. 125-138, [Link] -US. Department of Defense (2016) Unified Faciltes Criteria UFC) Design of Buildings to Resist Progressive Collapse (US. Department of Defense, UFC 4023-03) Viasss, AG, tzuddin, 8. A, Flghazoul A, ¥. andl Nethercot, OA (2008) Progressive collapse of mulstorey buildings due to sudden colurnn loss—Par Ik Application. Engineering Structures, 30(), 1424-1438 Wels, M. 2011) Tal timber buildings: Applications of soli timber con struction in multstory buleings. Counc on Tall Buildings and Urban Hebtat Journal, 2011(0, 26-27 Xiong, H. and Liu, ¥. (2016) Experimental study of the lateral resistance of bolted glam timber post and beam structural systems. Journal of Structural Engineering, 142/4), £4014002. YYazumura, M, Kobayashi, K, Okabe, M, Miyake, T. and Matsumoto, K (2016) Fullscae tests and numerical ana of low-sise CLT structures Under lateral loading. Journal of Structural Engineering, 142(0, 4015007.