10 © Ernst & Sohn Verlag für Architektur und technische Wissenschaften GmbH & Co. KG, Berlin · Steel Construction 5 (2012), No. 1 Articles This paper presents a new resistance model for steel plate gird- ers subjected to patch loading. The model aims to correct the EN 1993-1-5 formulation concerning the resistance to transverse forces with the help of findings presented in several doctoral studies on this topic. In addition, a statistical calibration of the new model is addressed. This calibration includes a new pro- posal for the resistance function χ-λ. The proposal is conceived for fulfilling safety levels of the partial safety factor that match the current values included in EN 1993 for instability-related prob- lems. These new proposals are in accordance with the design philosophy found in EN 1993-1-5 for the design of plated struc- tures. 1 Introduction The adequate and safe formulation of the patch loading resistance is crucial in the design of launched steel bridges. The resistance of plate girders subjected to patch loading has been studied thoroughly. A large number of contribu- tions concerning the web resistance to patch loading are available in the literature. Two milestone research works [1, 2] provide the framework for the formulation for the patch loading resistance presently implemented within EN 1993-1-5 [3]. In these works, the patch loading resis- tance is formulated by following a similar procedure found in other instability-based verifications, namely the χ-λ ap- proach. In this approach, plastic and elastic critical buck- ling resistances (F y and F cr respectively) are defined sepa- rately and subsequently related to the ultimate load by a resistance function χ= f(λ). Recent contributions [4–9] underpin the need for fur- ther modification of the plastic resistance F y which ap- pears in the current version of EN 1993-1-5. According to these studies, the definition of the plastic resistance over- estimates the patch loading capacity in certain cases (hy- brid girders), whereas this capacity is slightly underesti- mated for others (very slender girders). This paper provides a brief description of the modifi- cation needed. Furthermore, a design proposal including a new patch loading resistance model and a new resis- tance function – calibrated based on hundreds of experi- mental and numerical results – are outlined. 2 Current formulation in EN 1993-1-5 The verification of patch loading F Rd is based on simplifi- cations of the procedures provided in [1] and [2]. The gen- eral approach currently included in EN 1993-1-5 is based on a plastic resistance F y which is reduced by means of the resistance function (Eq. (1)). The plastic resistance in- cludes the length l y , which can be calculated from the geo- metrical and mechanical properties of the girders using Eqs. (2) and (3). Plate instability is accounted for by means of Eq. (4). The buckling coefficient k F is given in Eqs. (5), (6) and (7). This formulation can be applied even if the patch load is applied on only one side of the steel girder with the assumption that the flange is prevented from ro- tating about its longitudinal axis. The plastic resistance as well as the elastic critical buckling loads are both related to the ultimate load by means of the resistance function (Eq. (8)) and the relative slenderness (Eq. (9)). The formu- lation includes a partial safety factor γ M1 . (1) (2) (3) (4) (5) (6) (7) (8) F F f l t f at Rd F y M F yw y w M F yw w M = = χ γ χ γ χ γ · · · · · · · 1 1 1 l s t m m a y s f = + + + ( ) 2 1 1 2 · · m f b f t m h t if othe yf f yw w w f F 1 2 2 0 02 05 = = · · . · ., λ rwise m 2 0 = F k E t h cr F w w = 09 3 · · k h a F w = + 6 2 2 · k h a b a F w s = + + 6 2 5 44 0 21 2 1 · . . γ γ s sl w w w I h t a h b a = + 10 9 13 210 03 3 3 1 . · · · . χ λ F F = 05 10 . . Statistical evaluation of the new resistance model for steel plate girders subjected to patch loading Rolando Chacón* Benjamin Braun Ulrike Kuhlmann Enrique Mirambell DOI: 10.1002/stco.201200002 * Corresponding author: e-mail rolando.chacon@upc.edu