Comparative Efficiency and Performance of Glass Design Methods

This research compared two common glass design methods through a set of original worked examples on two-edge supported spans. A comparison of the results revealed the importance of accurately modeling actual material behavior to: (1) optimize materially efficient designs; and (2) produce designs with consistently reliable levels of risk. Results showed that minimization of material usage correlates to a consistent probability of failure metrics. For the two-edge supported spans (i.e., simply supported beam), the average monolithic plate thickness found using allowable stress design methods was 16.4% greater than that calculated using strictly probabilistic-based design.