HABILITATION

The methodological part of the research work provides an overview of the developed consistent mathematical approach for the optimization of life cycle costs of any complex, replaceable or repairable system under uncertainty. The essential contribution consists of the combination of condition and reliability theory with life-cycle approaches to a class of stochastic processes with continuous condition development and failure distribution. These processes allow the determination of stochastic time-dependent condition distributions and residual service lives for any continuous, scalable performance function and failure distribution. Including aspects of survival analysis, the simplifications and systematic deviations of deterministic as well as stochastic discreet approaches can largely be avoided. The application of the developed approach in condition prediction on network and asset level considers specific survey data allowing a mathematically sound assignment of condition related risks. The simultaneous modelling of performance functions for failure types, elements, and assets enables a specific allocation and optimization of measures with timing, length of work zones and availability at minimal costs. 

Applying the developed approach on road pavements „top-down“ allows an assessment of rehab and maintenance strategies as well as a determination of investments needs, remaining service life and residual asset value on network level based on a standardized life cycle. With Asset Management Standards and overview on existing Pavement Management Systems (PMS), a critical analysis of failure causes, condition thresholds, performance models and optimization approaches are presented. Implementing both standard approaches and the developed new approach provides compelling evidence regarding an insufficient use of funds due to methodical weaknesses of common PMS. By avoiding homogeneous sections, subjective weighting factors and efficiency optimization using economies of scale on treatment level these drawbacks can largely be avoided. However, the specific modelling of all variables leads to an extremely complex, nonlinear problem and an exponentially growing number of possible solutions with an increasing number of failure types, treatments and road sections. The developed unique algorithm allows for solving this problem for optimal treatment type, timing and work-zone length being capable of handling an unlimited number of road sections or road assets on a route. In addition, extensive comparative calculations „ceteris paribus“ show a substantial savings potential in contrast to common approaches. 

The application of this new approach to bridges and tunnels enables the derivation of a standardized lifecycle model as well as the determination of investment assets, financing requirements and availability at network level "top-down". Principal questions regarding the life cycle costs of planning alternatives can be answered as well at this level allowing to avoid inefficient investments and insufficient use of funds. Based on a comprehensive analysis of existing methods a comprehensive asset management approach for bridges and tunnels on element level "bottom-up" is provided as well. With standardized failure type and treatment catalogues as well as additional methodical considerations the gaps in existing approaches regarding a full Asset Management Cycle are closed. Apart from specific findings, the main benefits are the possibility to improve condition and reliability prediction based on actual measurements instead of unprecise grading. With the appropriate implementation of the presented concepts, the workload of road operators can be reduced freeing up resources for efficient monitoring and improved results. All relevant methodological foundations and results of this work are substantiated by theoretic and empirical evidence being published in high-level international peer-reviewed journals and conferences as well.