A comparison of deterministic and probabilistic life cycle cost analyses of ground source heat pump (GSHP) applications in hot and humid climate

Geothermal energy is widely considered as an ideal alternative energy for building operations. Despite the fact that ground source heat pumps (GSHPs) have been studied for decades, including economic analyses such as life cycle costing, their technical and economic applicability in hot and humid climate such as FL, USA is yet to be determined. In the past, many life cycle cost analyses on GSHPs were conducted using a deterministic method to derive point estimates. In many cases, data were assumed, but data uncertainties were not accounted for. In this study, a comparison of the deterministic and the probabilistic method was performed in order to understand the impact of data uncertainties on results of analyses. The probabilistic life cycle cost analysis was based on Monte Carlo simulation. A GSHP application case in Pensacola, FL was selected for the comparison. Some data were collected through site visits or interviews, while others were gathered from the literature or published data. Probability distribution functions used by Monte Carlo simulation were derived based on historical data or assumptions. Results of the case study from both deterministic and probabilistic methods confirmed that the GSHP option was economically more favorable than a conventional single zone split system using heat pumps, but with a long payback time if incentives were not considered. The probabilistic method was found to deliver a more reliable conclusion with more critical information than the deterministic method. However, further studies were needed to verify those initial observations obtained from this case study.

► Compared the deterministic and the probabilistic approach of life cycle costing.
► Incorporated uncertainties in life cycle costing.
► Enhanced reliability of conclusions by the probabilistic method.
► Incentives critical for the adoption of ground source heat pumps (GSHPs).