**8. Results and findings**

This chapter presents the results and findings of seven research papers that focus on the causes of bridge failures and methods to analyze bridge structures. Presenting frequency of bridge failures Wardhana looks at causes from design, construction, material, and maintenance. Zhang considers structure type and material with an emphasis evaluating structural strength and stability in a bridge's early life to monitor the rate of fatigue in structural elements. Evaluating arch bridges Trajber considers structure type and service age proposing repair and rehabilitation to reduce the rate of deterioration in arch bridges. For cable-stayed bridges Padic analyses the number of design parameters needed to account for the interaction of load-bearing structural elements. Pajeskirke evaluates balanced cantilever bridges in India used for hilly terrain with little bottom support which require special construction requirements. Arioglu uses regression models to examine design parameters for suspension bridges with application to the new 1915 Canakkale Bridge in Turkey. In looking at the I-880 freeway collapse in California in 1989 Mochle points out the failures was the result of not identifying the weakest point in the bridge, specifically the connection pins between the two decks. Analyzing the I-35 bridge collapse in Minnesota Salem uses the Applied Element Method (AEM) to accurately identify the exact gusset plate that failed. The I-35 bridge had long standing deficiencies.

The seven bridge failures presented this paper highlight a variety of causes that led to structural failures. In each of the seven failures there are valuable lessons to be learned. These lessons should be used by designers to build stronger and better bridges with longer service lives. Bridges should not fail.

By looking at bridge failures and their causes Heggade, VP of Board of Management of Gammon India Ltd., presents in his paper the fact there are valuable lessons to be learned in failures. He notes a majority of bridge failures occur in service without external action, during construction and in false works. He indicates there have been an alarming trend of bridge failures in Asian countries and discusses aspects of learning lessons in bridge failures from the Indian context. The point he makes is that accurate documentation of bridge failures is necessary for improved bridge designs. Study of failure improves design concepts for robustness, extrapolation, and durability. The study of bridge failures is an invaluable source of information on bridge design limitations. Bridge design is a process of anticipation of failure. Heggade states bridge designers must learn from past bridge failures to improve deigns to prevent bridge failure [14].

In recommending improved methods to reduce bridge failures Zhang points out:

"Researchers need to strengthen their research on the stability and fatigue of steel bridges, as well as inspection and maintenance. Extreme loads such as flood, collision, and overload contribute to a large number of bridge failures because of the lack of extreme loads data and design theory defects. It is critical for bridges to have

sufficient redundancy and capacity protection measures to reduce the probability of bridge failure due to extreme loads. Previous statistical methods and classification methods for the characteristics and causes of bridge failures lack unified standards, and a more scientific method needs to be established [8].

The key point Zhang makes is that unified standards with better scientific methods are needed to classify the characteristics and causes of bridge failures. More research is needed to significantly improve risks reduction of bridge structural failure.

One of the key findings to come out of the research for this paper is the need to better understand how and why bridge failures occur and to apply the lessons learned in failures to design and build better bridges, of all types, that will not fail.
