1989 年 402 巻 p. 121-131
A majority of structural failures are due to errors in planning, design, construction and utilization, rather than stochastic variability in material strengths and structural loads. Concerning to this fact, a study focusing on unexpected member failure occurring in relatively early stage of loading is carried out here to investigate statistically strength deterioration of structure due to this kind of member failure. Two types of double-layer grids-6×6 square plan truss (truss SQ) and 6×9 rectangle plan truss (truss RE)-are designed using a method described in Chap. 2. Two kind of safety margin value-α=1.2^6 (design f) and α=1.2 (design γ)-are considered. Elasto-plastic analysis assuming bi-linear member force-deformation relation is peformed to each intact truss and all the structures with one member removed. Statistical consideration on the rates of strength deterioraion due to various member removals is performed. It can be concluded as follows: 1) The strength deterioration rate due to member removal varies largely depending on the member safety margin of design. 2) With the same member safety margin, the strength deterioration rate due to member removal may vary largely depending on the method of choosing members for use. 3) For the trusses discussed here, removals of chord with large section, thus located in the center part of the truss induces large strength deterioration. 4) For the trusses discussed here, the removals of web member has small influence to the strength deterioration. 5) For truss RE designed here, the strength deterioration caused by the removals of short-side chord is much larger than that caused by the removals of long-side chord. This can be improved by reinforcing the long-side chords.