Abstract— The paper presents quantitative RAM-based economic models capable of sup-plying early and realistic information on the envisaged lifespan economic perfor-mance of energy infrastructure systems during their conceptual designs. Unlike the conventional cost models which do not take these RAM characteristics into account, the models in this paper provide an early trade-off between the RAM attributes the designer wants to incorporate in the system and the cost of having them. The economic models thus provide the designer the ultimate leeway of selecting more intrinsically reliable and more inherently maintainable components or subsystem of the infrastructure system at a more affordable investment costs. The paper focusses on a user friendly formulation for incorporating the social costs, which are typical for infrastructures that operate in the public domain, in the cost structure of infrastructure systems. To demonstrate the implementation and utility of these economic models, it has been applied to a District Heating Network (DHN) case study. I. INTRODUCTION One of the key aspects of infrastructure systems is their relatively long life span, e.g., several decades. Over this long a time span it is not only that the value of money significantly changes, but also the income from delivered services and goods can change. Furthermore, in keeping up the performance of the infrastructure, several re-investments in new technologies (retrofits) may be necessitated. These require that a life span perspective must be taken in setting up the economic model. Therefore, in most of the cost formulation the discrete lifespan perspective of costs and benefits is the main mathematical structure of the economic model. What is being emphasized in this paper is the fact that the RAM (Reliability, Availability, Maintainability) as well as the economy of the infrastructure system (as infrastructural performance indicators) should be consecutively carried along the length and breadth of the concept and early design phases as well as in the entire life cycle. The overriding importance of integrating such economic performance indicator that takes into account RAM related strategic issues in these specific phases stems from the fact that a very high proportion of the potential to influence life cycle cost and revenue generation of the infrastructure systems is fixed in these phases. This has been extensively discussed in [1][2][3]. Apart from the life cycle perspective, a novel cost formulation that adds the social cost of infrastructure service outages to the RAM-enriched cost equations is the hallmark of this paper. The next section briefly introduces the Markov RAM-based economic models and section III will build upon this model by introducing cost models that reflect the social outages. In section IV, the applicability of the models are demonstrated by means of a case study of a district heating infrastructure design. II. MARKOV RAM-BASED ECONOMIC MODEL By Markov RAM-based economic model, we mean a set of mathematical relationships between cost and system design variables, as well as other performance metrics like RAM, which could realistically predict cost and other economic indicators from the values of these variables and metrics. For instance, from the availability models, the downtime at any given performance level due to failures could be determined and the associated costs and e orts of having such availability as well as the revenue loss accruable to it can be fed into the economic model. Figure 1 Graphical illustration of the cost and revenue components of a large scale infrastructure system In Figure 1, the various cost and revenue components as well as the stages in the life cycle of the infrastructure system where such costs and revenues are incurred or accrued are depicted. These costs, especially those incurred at the operations and maintenance stage as well as construction and decommissioning stage, form the major elements of the economic model that have been developed. We refer to [3] for detailed derivations of the various elements of our cost models. The overall stochastic model comprises the following elements: - social (user) cost model (see next section) Integration of societal outage cost into infrastructure design and maintenance optimisation Paulien M. Herder, Member, IEEE, Austine N. Ajah, Margot P.C. Weijnen, Member, IEEE Delft University of Technology / Next Generation Infrastructures Foundation PO Box 5015, 2600 GA Delft, The Netherlands, p.m.herder@tudelft.nl Proceedings of the 2009 IEEE International Conference on Systems, Man, and Cybernetics San Antonio, TX, USA - October 2009 978-1-4244-2794-9/09/$25.00 ©2009 IEEE 4257