Identifying the Pareto-front approximation for negotiations of BOT contracts with a multi-objective genetic algorithm

The build-operate-transfer (BOT) project is formulated based upon cooperative partnerships among various public and private parties. Win-win negotiation solutions are especially essential for parties that usually have conflicts of interests. A simulation-based multi-objective genetic algorithm (SMOGA) procedure is developed to address conflicts among the objectives of three parties: government and consumer interests that focus on users' rights and social welfare; the private investor's pursuit of profit maximization; and lenders who want to control the solvency risk of the concessionaire. Taking the Kaohsiung cable car project in Taiwan as an example, the SMOGA procedure is used to seek terms and conditions lying on or near the Pareto frontier for BOT projects with multi-objectives. Monte Carlo simulation is conducted first to determine the project risks and generate the sample data used as the input of the genetic algorithm (GA). The GA subsequently generates a number of combinations of concessional terms and conditions that approximately and simultaneously optimize the three conflicting objectives. The GA-generated approximate Pareto frontier provides a decision space of financial and concessional terms and conditions that satisfies all parties and facilitates the negotiations of BOT contracts.