2 



1 



J = 1 



dj Cji,; k = 



to ei- = dci,; k = 



where d is defined as the system dependability. 



The uncertainty associated with estimating system dependability, d, 

 can be considerably less than first appears possible. Uncertainty, though 

 significant, is typically confined to a small number of elements. For 

 example, while there is little information on the reliability of multicable 

 systems used for heavy lift, the reliabilities of associated winches, ships, and 

 positioning systems are estimable with a fair degree of certainty. The less 

 certain factors of cable life and the problem of tangling, for instance, can be 

 assigned a range of values. All other assumptions concerning the system 

 remain constant, and the results thereby computed will provide estimates of 

 the sensitivity of the system (financial or otherwise) to changes in the value 

 of the uncertain technical or operational parameter. 



In this (conceptual) stage of the project, the system dependability, 

 d, is the equivalent of conventionality. Consequently, the more "dependable" 

 system is the more conventional system. Radical departures from the norm 

 cannot realistically be assigned high levels of dependability because of the 

 inherent lack of favorable evidence. Critical decision-making in design is 

 valid only if the level of confidence is based primarily on the available 

 experimental and operational evidence, with some but less emphasis placed 

 on theory and intuition. When it is stated that in the context of this study 

 that System A is more dependable than System B, the inference is that the 

 available evidence indicates the former has (at least) a smaller number of 

 unsolved problems, i.e., System A is more conventional than System B and, 

 therefore, is a more dependable system. Implicit in this definition of d is the 

 assumption that any candidate is physically and operationally feasible if 

 indefinite amounts of time, money, and effort are alloted to its development. 

 However, the constraints of time and money and a sense of what is reasonable 

 have forced the authors to rely heavily on logical extensions of the state-of- 

 the-art. The narrow time frame for this project, i.e., a 600-ton lift capacity 

 by 1977, leaves no other alternative. Moreover, the priority given the heavy- 

 lift project, or for that matter any other ocean engineering project, does not 

 lead one to assume that more than average development funding will be 



37 



