No abstract

Abstract:As environmental and energy issues continue to emerge as global concerns, Leadership in Energy and Environmental Design (LEED) certification is becoming highly valued. However, since additional costs for LEED certification cannot be estimated before proceeding with certification projects, financial losses are often incurred. Additional construction costs are the most significant issue faced by enterprises aiming for LEED certification. Rough estimates of the range of additional construction costs are available, but it is difficult to identify factors that increase or decrease the price of a building. Thus, there is a need for a program that provides average data for LEED certification costs and suggests the easiest way to attain credits for a building. Considering that LEED certification is a rating system, this study develops an optimization algorithm that aims to derive the minimum score for a desired LEED level at minimal cost. Credits are studied and classified by their difficulty and the required cost, allowing for an algorithm that can suggest a customized approach to acquire the minimal required score. The practical, data-driven program developed herein helps shorten the consulting process and increases the accessibility of LEED certification.

No abstract

The fatigue life prediction of steel bridges depends on the amplitude, stress ratio, stress interaction (stress order), and frequency of variable stress spectra. The stress interaction is always altering the crack growth rate under the application of variable stress spectra. The random nature of variable stress spectra makes it difficult to model all these influential parameters correctly. In this study, the fatigue prediction model of steel bridges based on linear elastic fracture mechanics was developed by using measured variable stress spectra of three steel bridges. Variable stress spectra data are used to calculate and analyze the entropy index of stress interaction and crack severity index (CSI) of the effective stress associated with the characteristics of the variable stress spectra. Using these results, the results of fatigue analysis, and multiple regression, we develop prediction models for entropy index-CSI-fatigue crack length, average daily truck traffic (ADTT)-CSI-fatigue crack length, and ADTT-entropy-fatigue crack length. These models can predict the fatigue life of a bridge structure, and this information can be used to plan lifetime maintenance for similar bridges with variable stress spectra.

This paper presents the simple approaches to undertake reliability-based design optimization (RBDO) of bridges by integrating the first order reliability approximation and optimal algorithms of Matlab optimization toolbox. The approaches are applied to solve three RBDO problems that represent almost the practical designs of bridges. At every step of the optimal algorithm, reliability index is calculated without increasing the number of constraints to reduce the computational cost. Moreover, these approaches allow users to formulate the real RBDO problems with the separation or combination of component and system failure probabilities as well as equality and inequality probabilistic constraints. Numerical examples indicate the improvement of approaches due to lower design variables and lower cost for comparing to existing results with the same value of target reliability index. The proposed approaches are simple but efficient for solving RBDO problems to produce the better results than existing approaches and the obtained results may be closer to the exact optimization results.