Lebenszyklusmanagement von Brücken mit Hilfe von Key Performance Indikatoren auf Basis unterschiedlicher Erfassungsquellen

The maintenance of existing bridges poses major challenges for road authorities. They have to ensure the continued operation of these crucial infrastructure components, despite a permanent increase in traffic and a simultane-ous deterioration in the condition of the structure with time. Under these circumstances the correct prioritization of the necessary maintenance measures is essential. Various parameters have been introduced to quantitatively describe the condition of a structure in order to assist such an assessment. These in turn can be summarized in so-called key performance indicators (KPI), which enable a cumulative, holistic assessment of the structure. Three es-sential, internationally recognized KPIs are reliability, availability and safety.
In the research project, various methods of condition assessment, such as non-destructive testing or monitoring methods, were gathered and then systematically analyzed. Various categories of information on structures were formed, focusing on the ways they can be integrated into the structural assessment. Based on logical causal relationships and 30 case studies for monitoring measures or investigations using non-destructive testing methods on bridges in Germany, a link was created between the methods of condition assessment and the available information on the structure. This enabled the next step of the integration of the available structural information into the condition assessment with the three KPIs examined in the project. This was more complex for reliability than for safety, as a distinction had to be made between the calculated load-bearing capacity and the observed condition of the structure, which were thereafter combined into an overall assessment.
Availability plays a particularly important role in the KPIs for bridges. It is only achieved if both reliability and safety reach a certain required level. It can therefore not be considered independently of reliability and safety, but is rather a result of these two. This is precisely why it is one of the most important KPIs, as the importance of reliability and safety primarily stems from the fact that they are the prerequisite for availability in the long term. At the same time, we are dependent on forecasts for the future development of the other two KPIs to determine the upcoming availability. For this reason, the research project focused extensively on AI-supported forecasting methods of condition assessment indicators. Various approaches were investigated, particularly with regard to the clustering of structures that are used as the basis for a forecast.
Finally, an important aim of the research project was not only to describe the developed approaches theoretically, but also to demonstrate their fitness for practical implementation. For this reason, a software tool was developed in parallel that capable of importing the structural data from the “SIB-Bauwerke” software and of implementing the developed processes and algorithms. For this purpose, six bridges, for which additional information is available, were selected as demonstrators and the integration of the provided additional data into the KPIs was demonstrated in the software.
The result of the project is thus a tool with which the three KPIs of reliability, safety and availability can be determined for a structure with heterogeneous data. The data available in “SIB-Bauwerke” always serves as the starting point. The bridge operator can then systematically categorize the different methods of condition assessment at his or her disposal according to the type of structural information that is gained and include it in the assessment. The software uses complex, AI-supported algorithms to determine forecasts that are reflected in estimates of future availability. These indicators can then support road administration offices in their decisions when drawing up maintenance strategies and prioritizing investments with a limited budget. The KPIs can thus contribute signifi-cantly to an information-based and responsible life cycle management when dealing with existing bridges.