Time-cost trade-off (TCTO) is one of the viable techniques in project management. Finding nearoptimum solution is a very important goal that received a great deal of interest. TCTO problems have been traditionally solved by two distinctive approaches: heuristic methods and optimization techniques. Although heuristic methods can handle large-size projects, they do not guarantee optimal solutions. In this paper, an optimization model is developed employing genetic algorithm technique for TCTO problem that minimizes project direct cost and takes into account discounted as well as net cash flow with the least possible duration. Costs of activities are assumed to be incurred at their finish times, uniformly distributed or following S-curve pattern. The model provides near-optimal solution in which precise discrete activity time-cost function is used. The model input includes precedence relationship between project activities, discrete utility data for project activities associated with cash-in and -out. Model formulation and implementation are deliberately delineated, verified and validated then implemented to several case studies. The predictions of the proposed model demonstrate that selected activities' durations as well as costs and consequently optimal project duration significantly differ from traditional analysis if only conventional cash flow is considered. The model provides project management engineers and practitioners with a rigorous tool for considering the net cash flow in time-cost decisions so that the best alternatives can be identified.