| Brief Description of Research Project | The objective of this study is to develop a data-driven approach to evaluate conditions of small and medium span bridges by proposing new predictive models based on the existing FHWA bridge inspection and maintenance data using 5,333 bridges in Maryland. Data includes age of the bridge, AADT, ADTT, material, structural system, length, width, and weather condition. The bridges will be categorized based on their type and kind, for each bridge from each category, the historical data will be collected, data processing, cleaning and analysis will be performed, and several machine learning algorithms will be used to develop predictive models that determines bridge condition rating. Using that model with highest accuracy, the deterioration of the bridge and its rating will be forecasted for the next 20 years from the day that data was extracted, and the model can be updated as new FHWA data becomes available. The proposed tool can help decision makers to accurately predict bridge condition and effectively allocate funds for the maintenance, rehabilitation, and repair of bridges. |
