Gregory J. White

Life expectancy assessment of marine structures

Abstract In this paper, a methodology of structural life assessment is suggested. The methodology is based on probabilistic analysis, using reliability concepts and the statistics of extremes. The methodology results in the probability of failure of a structural system according to the identified failure modes as a function of time, i.e. structural life. The results can be interpreted as the cumulative probability distribution function (CDF) of structural life. Due to the unknown level of statistical correlation between the safety margins of components of the system, limits or bounds on the CDF of the structural life can be established. The limits correspond to the extreme cases of fully correlated and uncorrelated safety margins. The effect of inspection strategies on structural life is discussed. An example illustrating the use of the methodology is presented.

Code Development for Ship Structures—A Demonstration

A demonstration summary of a reliability-based structural design code for ships is presented for two ship types: a cruiser and a tanker. One reason for the development of such a code is to provide specifications which produce ship structure having a weight savings and/or improvement in reliability relative to structure designed by traditional methods. Another reason is to provide uniform safety margin for ships within each type. For both ship types, code requirements cover four failure modes: hull girder bulkling, unstiffened plate yielding and buckling, stiffened plate buckling, and fatigue of critical detail. Both serviceability and ultimate limit states are considered. Because of limitation on the length, only hull girder modes are presented in this paper. Code requirements for other modes will be presented in future publication. A specific provision of the code will be safety check expression, which, for example, for three bending moments (still water Ms , wave Mw , and dynamic Md ), and strength Mu , might have the form, following the partial safety factor format: γsMs + γwMw + γdMd ≤ φMu γs , γw , γd , and φ are the partial safety factors. The design variables (M ’s) are to be taken at their nominal values, typically values in the safe side of the respective distributions. Other safety check expressions for hull girder failure that include load combination factors, as well as consequence of failure factors, are considered. This paper provides a summary of safety check expressions for the hull girder modes.

Applications in Ship Structures

The structural weight of naval surface combatants constitutes approximately 35% of the total lightship displacement, making hull structures the heaviest of all ship subsystems. Any improvements in vessel capability through growth of the mission-related payload will necessitate an equivalent reduction in weight in some other subsystem. Because of the proportionally low cost of the hull structure subsystem, as shown in Fig. 24-1, improvements can be made here without drastically increasing total vessel cost.

Probability-based life Prediction

The estimation of the life expectancy of a structure is not a simple task. Many factors affect the life of a structure. These factors include design parameters, design safety factors, design methods, type of structure, structural details, materials, construction methods and quality, loads, maintenance practices, inspection methods, and other environmental factors. These factors have different types of uncertainty associated with them. Generally, these factors have the following types of uncertainty: (1) physical randomness in magnitude and time of occurrence, (2) statistical uncertainties due to using limited amount of information in estimating the characteristics of the population, (3) model uncertainties due to approximations in the prediction models, and (4) vagueness in the definition of various factors and their effect on life. Therefore, the estimation of life expectancy is a complex process.

Uncertainties in resistance and strength measures of marine structures

To develop reliability-based design formats for marine structures, the uncertainties associated with the strength side of the various limit-states need to be investigated. A methodology is presented for quantifying the uncertainty in the strength measures and the analytical models used in the design of marine structures. Existing design codes in other fields of engineering are reviewed for limit-state expressions and measures of uncertainty in modeling. One of the limit-states is chosen as an example of an approach for determining the level of uncertainty associated with the strength parameters and the analytical model.<<ETX>>

Life Expectancy Assessment of Structural Systems

The assessment of structural life expectancy for marine vessels is a relatively complex task. This is due to uncertainties in the various parameters as well as incomplete knowledge of their characteristics. In this a paper, a methodology for structural life expectancy is suggested. The methodology is based on structural reliability, theory of extremes, a plate wastage model, and system analysis. Example applications using marine structures are discussed.

Life expectancy of hull structures of boats

A methodology for the structural life assessment of a ships structure is suggested. The methodology is based on probabilistic analysis using reliability concepts and the statistics of extremes. In this approach, the estimation of structural life expectancy is based on selected failure modes. For the purpose of illustration one failure mode is considered in this study. This is plate plastic deformation. Structural life, based on this failure mode, for an example vessel is determined.

Load and Resistance Factors for Structural Components

ABSTRACT Recent investigations have shown that Probability-based design formats using the First-Order Second-Moment (FOSM) Method and the Advanced Second-Moment (ASM) Method may result in engineering designs of different reliability levels than the ones specified in developing the design format. This difficulty is due to the manner in which the methods select the Most Likely Failure Point. In this paper a formal definition of the Most Likely Failure Point is given. The ability of the FOSM and ASM Methods, and the new Reliability-Conditioned (RC) Method to arrive at this Most Likely Failure Point are evaluated.