Abstract
During planning of repetitive build-operate-transfer (BOT) construction projects, investors and (or) investor-representatives are always faced with a challenging task to identify an optimal project plan that maximizes the investor's profit within a specified concession period. Minimizing project duration in BOT projects often results in additional costs, however, it increases the project operation period and accordingly increases the investor's profit. Optimal plan of constructing repetitive BOT projects requires identifying construction methods and the associated start and finish time of each construction activity according to the availability of resources and contractors. Traditional scheduling methods such as critical path method (CPM) networks do not provide efficient planning of repetitive construction projects as they (1) do not maintain crew work continuity as well as resources availability, and (2) often result in large and repetitive number of activities to model repetitive construction projects. This paper presents the development of an optimization model using linear scheduling method that identifies the near-optimal plan for constructing repetitive BOT projects to maximize the investor's profit. The development of this optimization model includes three steps: formulation phase that formulates the model decision variables, objective function, and constraints; implementation phase that executes the model computations using genetic algorithms and identifies the model inputs and outputs; and evaluation phase that verifies the model performance and document its value. A case study of a construction project is presented to illustrate the new capabilities of the model. The results of the model showed 21% profit increase as compared to the existing schedule adopted by the investor.