Uncertainty in life cycle costing for long-range infrastructure. Part II: guidance and suitability of applied methods to address uncertainty

Abstract

Life cycle costing (LCC) is the state-of-the-art method to economically evaluate long-term projects over their life spans. However, uncertainty in long-range planning raises concerns about LCC results. In Part I of this series, we developed a holistic framework of the different types of uncertainty in infrastructure LCCs. We also collected methods to address these uncertainties. The aim of Part II is to evaluate the suitability of methods to cope with uncertainty in LCC. Part I addressed two research gaps. It presented a systematic collection of uncertainties and methods in LCC and, furthermore, provided a holistic categorization of both. However, Part I also raised new issues. First, a combined analysis of sources and methods is still outstanding. Such an investigation would reveal the suitability of different methods to address a certain type of uncertainty. Second, what has not been assessed so far is what types of uncertainty are insufficiently addressed in LCC. This would be a feature to improve accuracy of LCC results within LCC, by suggesting options to better cope with uncertainty. To address these research gaps, we conducted a systematic literature review. Part II analyzed the suitability of methods to address uncertainties. The suitability depends on data availability, type of data (tangible, intangible, random, non-random), screened hotspots, and tested modeling specifications. We identified types of uncertainties and methods that have been insufficiently addressed. The methods include probabilistic modeling such as design of experiment or subset simulation and evolutionary algorithm and Bayesian modeling such as the Bayesian latent Markov decision process. Subsequently, we evaluated learning potential from other life cycle assessment (LCA) and life cycle sustainability assessment (LCSA). This analysis revealed 28 possible applications that have not yet been used in LCC. Lastly, we developed best practices for LCC practitioners. This systematic review complements prior research on uncertainty in LCC for infrastructure, as laid out in Part I. Part II concludes that all relevant methods to address uncertainty are currently applied in LCC. Yet, the level of application is different. Moreover, not all methods are equally suited to address different categories of uncertainty. This review offers guidance on what to do for each source and type of uncertainty. It illustrates how methods can address both based on current practice in LCC, LCA, and LCSA. The findings of Part II encourage a dialog between practitioners of LCC, LCA, and LCSA to advance research and practice in uncertainty analysis.