Solar chimney power plants have been accepted as one of the promising technologies for solar energy utilization. The objective of this study is to propose an effective approach to simultaneously determine the optimal dimensions of the solar chimney power plant and the economic feasibility of the proposed plant. For this purpose, a two-stage economic feasibility approach is proposed based on a new nonlinear programming model. In the first stage, the proposed optimization model which determines the optimal plant dimensions that not only minimize the discounted total cost of the system, but also satisfy the energy demand within a specified reliability taking into account the stochasticity of solar radiation and ambient temperature is solved using a commercial optimization solver that guarantees finding the global optimum. In the second stage, the net present value of building the plant is computed by deducting the discounted total cost found in the first stage from the present value of revenues obtained due to selling the electricity generated by the plant. The proposed approach is novel because it determines the optimal dimensions of the plant together with its economic feasibility by taking into account the energy demand and uncertainty in solar radiation and ambient temperature. The proposed approach is applied on a study in Potiskum, Nigeria, which reveals that building a plant with a collector diameter of 1128 m and chimney height of 715 m to Potiskum would be profitable for investors at an annual rate of return of 3% and would provide electrification to about 7500 people with a high level of reliability. The proposed approach is benchmarked with an intuitive approach and an approach that does not consider uncertainty in solar radiation and ambient temperature. The results clearly revealed the value of the proposed approach. Managerial insights on the impact of the efficiency of the collector, the efficiency of the turbine, electricity price, electricity demand, meteorological conditions, and discount rate on the size of the plant and the net present value are obtained through detailed sensitivity analyses. (C) 2016 Elsevier Ltd. All rights reserved.