COST-BENEFIT ANALYSIS (CBA) OF ENERGY RETROFIT IN SINGAPORE INSTITUTIONAL BUILDINGS USING IES-VE ENERGY SIMULATION MODEL � A CASE STUDY OF NUS SDE 2

Abstract

To have at least 80% of buildings certified green by year 2030, energy retrofits on existing buildings are essential. Although building energy simulation models can estimate energy savings, cost-benefit viabilities of potential retrofit alternatives on institutional buildings often remain unclear. Thus, this paper proposes a framework for optimal decision making of energy retrofit in existing institutional buildings using energy simulation model. NUS SDE 2 is selected as a case study to contextualise the analysis with IES-VE model.

The framework has integrated several aspects of energy retrofit, particularly model calibration, data clearance and decision making. Calibrated with observational, experimental and energy audit data to reflect effects of occupancy and non-occupancy based changes, Mean Bias Error (MBE) and Coefficient of Variation of the Root Mean Square Error [CV (RMSE)] of the model are established as 3.942% and 8.570% respectively, which are within the most stringent values. Next, potential retrofit options and estimated energy savings are identified using hybrid approach of model simulations and heuristic method. The collected material unit cost is adjusted to project unit cost in the Singapore context depending on 3 increasing levels of retrofit with rates of 2.520, 2.828 and 3.174 respectively. Then, Cost-Benefit Analysis (CBA) is used to appraise viabilities of different retrofit options, which includes Net Present Value (NPV), Benefit Cost Ratio (BCR) and Payback Period (PP). After sensitivity analysis, decision matrix is used to compare different alternatives using multiple criteria and weightages for optimal decision making. The retrofit combination that aims to improve occupant behaviour for energy savings has topped with a score of 99.42, which is 46% better than the rest.

In short, the proposed framework has achieved multiple objectives of model calibration, retrofit options identification, cost data clearance, CBA and retrofit decision optimisation. By overcoming the limitations in data validation, framework application and utility extension, future studies can focus on generalising and extending the framework to encompass more building types and locations for wider adoption.