This work proposes two algorithms for maximizing the sum-rate and fairness in downlink non-orthogonal multiple-access (DL-NOMA) systems with indoor-outdoor user pairing. Specifically, one of the proposed algorithms is based on the streamlined simplex method (SSM), while the other utilizes the least and most cost method (LMCM). The fairness rate in both cases is optimized using the max-min approach for the indoor device, which is assumed to have a lower channel gain due to its inherently more challenging environment. Various performance metrics such as the achievable data rate, fairness index, and energy-efficiency are analyzed to evaluate the effectiveness of the proposed algorithms relative to different benchmarks such as the Near-Far pairing (NFP) and the random user-pairing (RUP). The results show that the LMCM algorithm incurs lower complexity and offers better data rate and energy-efficiency when compared to the SSM algorithm, the NFP and RUP methods.