This thesis presents a multi-criteria approach to capacity expansion planning of electric utilities. Traditional least-cost generation expansion planning has become inadequate due to the prevailing multiple, conflicting objectives such as cost, enviromental degardation, and nuclear hazared. This is particularly true with emerging concerns over carbon dioxide emissions that is believed to contribute to global warming. In this thesis, we present the preference order dynamic programming approach, so that this new logic can be implemented within the already available dynamic programming based capacity expansion planning tool, called WASP. This approach is well-suited to examining utility planning issues such as generation mix, reserve margin, reliability, government regulation, and regional and global environmental issues. Through our case study, we note the importance of considering global warming as well as nuclear hazards. We also note that substituting plants that use cleaner fuels such as natural gas for plants tnat use carbon intensive fuels such as coal, is more effective in controlling carbon dioxide emissions. This substitution method is also more effective than replacing these plants with carbon dioxide-free nuclear units. By utilizing the multi-criteria model, we are also able to find the strategy for establishing long term power development plan satisfying carbon dioxide emission constraint. The solution can conform to the global greenhouse gas reduction strategies of the electric power sector in meeting the regional and golbal environmental protection objectives.