۱- Introduction

۲- Related work

۳- Methodology

۴- Case study

۵- Conclusions

References

Abstract

Civil infrastructure system maintenance planning is to determine which facility should be repaired, when and how maintenance should be carried out, and what treatment should be used under budget and other resource constraints. In the existing literature, various simulation and optimization models have been developed to help select the optimal maintenance plan. However, the developed models overlooked the deterioration propagation between adjacent connected facilities of the network infrastructure system. For instance, a facility receiving zero maintenance or having a failure of maintenance treatment affects not only the condition of itself, but also the deterioration rate of its neighboring facilities. This raises the call for taking the deterioration propagation into consideration when developing optimization models and capture to which extent it can affect the optimal maintenance plan. Therefore, in this paper, an infrastructure maintenance planning model considering the deterioration propagation between facilities is formulated as a mixed integer linear programming problem. A heuristic algorithm was proposed to solve the problem efficiently. Example networks were tested for the performance comparison between CPLEX and the heuristic algorithm. The proposed model performs better than models without the deterioration propagation.

Introduction

Civil infrastructure systems (e.g., roads, bridges, water supply, and wastewater) are exposed to aging effects and eventually subject to failure if no maintenance intervention is carried out. To prevent/delay failures, maintenance treatments need to be applied periodically. In order to optimize the allocation of resources for maintenance of the facilities, maintenance planning is needed. The primary objective of maintenance planning is to help decision makers schedule maintenance actions and determine which facility need to be maintained when maintenance should be carried out and which treatments should be used. As transportation infrastructure systems are spatially distributed assets covering large regions, they are specially characterized by the interdependence and interaction within and between different systems. Researchers have developed maintenance planning models addressing different relationships between facilities/systems from functional, economic, and other perspectives. For example, Bernhardt and McNeil (2004) stated that pavements are interconnected through geography, which implied the economies of scale in contracting long stretches of pavement for rehabilitation and the diseconomies of scale in terms of the disruption to users. Gao and Zhang (2013b) also pointed out that road sections selected for maintenance by traditional optimization approach are usually distributed spatially across the network. The authors suggested that, to take advantage of economies of scale, adjacent pavement sections with similar maintenance needs should be maintained within a single project. Rasmekomen and Parlikad (2013) conducted a study on optimizing maintenance plans for industrial assets by considering degradation and performance interaction between them.