A method for computing the minimum active power loss in competitive electric power markets is proposed. The active power loss minimisation problem is formulated as an optimal power flow (OPF) with equality and inequality nonlinear constraints which take into account the power system security. The OPF has been solved using the multiple predictor-corrector interior-point method (MPC) of the family of higher-order interior-point methods, enhanced with a procedure for step-length computation during Newton iterations. The utilisation of the proposed enhanced MPC leads to convergence with a smaller number of iterations and better computational times than some results reported in the literature. An efficient computation of the primal and dual step-sizes is capable of reducing the primal and dual objective function errors, respectively, assuring continuously decreasing errors during the iterations of the interior-point method procedure. The proposed method has been simulated for several IEEE test systems and two real systems including a 464 bus configuration of the interconnected Peruvian power system, and a 2256 bus scenario of the South-Southeast interconnected Brazilian system. Results of the tests have shown that the convergence is facilitated and the number of iterations may be small.