Light-frame wood roof assemblies may be stronger than what the current
single member design approach suggests because of repetitive member l
oad sharing and partial composite action. These strengthening effects
are partially addressed in the National Design Specification for Wood
Construction through a repetitive use factor of 1.15, applied to the a
llowable bending stress for repetitive wood members. This paper examin
es the effect of load sharing and composite action on wood pitched-roo
f assemblies through a new analysis model. Individual truss and roof a
ssembly strength distributions were computed using the new model. The
results show that slope, span, and other truss characteristics cause s
ignificant changes in the system factor for a given roof structure con
figuration. Computed system factors are presented in terms suitable fo
r working stress and reliability-based design methodologies. Computed
working stress design system factors vary from 1.15 to over 1.50 for t
he roof structures studied, suggesting that the current 1.15 factor ma
y be conservative for many roof structures. This conservatism must be
retained until more sophisticated analysis models, such as the one pro
posed herein, are fully verified and accepted for design.