A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering, School of Environment, Resource and Development

Thesis (M.Eng.) - Asian Institute of Technology, 1999

In order to solve the Municipal Solid Waste problem in Taiwan, Taiwanese Environmental Protection Agency (EPA) decided to execute the development of Refuse-to-Energy (RTE) Plant in privatization options such as BOT and BOO. However, because of without enough experience of BOT/BOO project in Taiwan, it is so often expected that high uncertainties can be included in estimates and forecasts of the total project. Therefore, the total risk of the project in the context of BOT/BOO procurement can be considerably higher than the risk of other public projects that are procured under the public finance and general project procurement methods. Meanwhile project sponsors who are responsible for managing, controlling, and affording/absorbing of project can face risk at several stages due to their failure to evaluate appropriately the project risks. It is necessary to find solutions in a comprehensive manner In this research, questionnaire is prepared to identify the major factors as well as the occurrence probability of major factors including uncertainty and then it is distributed to ten categories of professions. Factor Analysis Method is used to choose the risk factors which are ranked in high relative importance on either BOT or BOO optioned projects. One probabilistic model is used to gauge the combined effects of multiple uncertainties as well as to provide a consistent basis for comparing projects on different options. Risk-based cash flow analysis is used to quantitatively assess the difference between development options such as BOO/ BOT and different design conditions. Internal Rate of Return (IRR) is selected as the indication of economic effectiveness in BOO/BOT project. In development of probabilistic model, first of all, the probability densities of the uncertain parameters which are obtained from the experts through questionnaire survey are determined. From this qualitative approach, the stochastic model of Degree of Recurrence Risk (DRR) in RTE system could be developed in which three major categories such as social, environmental and financial are explained by taking into account of planning, construction and operation/maintenance stages. Through Monte Carlo Simulation approach, which is incorporated to the model of total DRR on Taiwanese Refuse-to-Energy system development, it is found that BOO would create a little bit higher recurrence risk than BOT. Since the developer with BOO option to procure RTE project has to invest huge budget to secure the land for facility as well as to conduct EIA for corresponding project which don't be required for BOT option. It is revealed that the environmental category would have the highest DRR and, in stages, the planning shows the highest DRR level. By reflecting the obtained result from risk-based cash flow, it is concluded that BOO option would create higher IRR but it would make higher risk than BOT option. It is also found that fixed value of designed variables shows lower risk than other cases. And the case with high growth rate of land cost even shows higher IRR value but the risk is lower than other cases. Another result is the change of DHV has a significant impact on IRR value in fixed O&M cost than the result from fixed DHV (1,800 Kcal/kg).