Partial and total factors of safety in anchored sheet pile design: Discussion

Abstract

Valsangkar and Schriver have contributed to furthering the use of limit state design methods in geotechnical engineering. The title of the paper implies partial safety factors for anchored bulkhead design, but the only limit state that was examined was kickout. Design of the bulkhead requires examination of a number of other limit states including anchor rod capacity, sheet pile yield, wale yield, and anchorage capacity (Navfac 1986), and all of these use soil and water pressures in their formulation. It would be desirable to use, for example, the same resistance factor for the soil friction angle for each limit case. To establish the adequacy of the recommended partial safety factors, the other limit states for an anchored bulkhead should also be compared with conventional design. An interesting outcome of this paper is that with the proposed partial safety factors for limit state design, the sequence of solving of statics problem, i.e., how water pressures are introduced into the moment equation, affects the design. The authors recommend that commentary be included in the Canadian Foundation Engineering Manual (Canadian Geotechnical Society 1985) to clarify this point. The partial safety factors discussed by the authors are applied to the loads and resistances in the design and are used to account for uncertainties in these values. However, implicit in these factors are components that account for uncertainties in the dimensions and method of analyses used. Water has no shear strength and virtually no variability of unit weight. The quantity that causes variability in the water pressures is the water level difference, i.e., a dimension. In some cases, such as this water level difference, should a separate factor be considered that would be applied directly to the dimension before being used in the limit state equation? If applied in this case, the sequence of incorporating water pressures in the moment equation would not change the design. For anchored bulkheads in particular, another dimension that should be considered for a separate factor is the dredge depth in front of the bulkhead. Dredging, which often occurs on a regular basis during the life of a bulkhead, may give a higher variability to the passive pressure than would normally be experienced with passive pressure being used in other types of design. Terzaghi (1954), Tschebotarioff (1973), and Navfac (1986), among others, recommend an increase in the embedment depth to account for this. Since this type of partial safety factor is thus used by many in conventional design, more widespread use of a partial safety factor applied to dimensions in limit state design should perhaps be considered.