This article presents a mixed integer linear programming model for the problem of operation planning of an electrical distribution system. The model defines the most appropriate adjustment of a set of control variables to minimize active energy losses of the system. The generation of active and reactive powers of distributed generators, the number of modules in operation of capacitor banks, and the number of tap steps for the voltage regulators are considered as the control variables. In the proposed formulation, the steady-state operation of the electrical distribution network is modeled mathematically, using linear expressions. The use of a mixed-integer linear model guarantees convergence to optimality using existing optimization software. The model was implemented in the mathematical modeling language AMPL and solved using the commercial solver CPLEX. A test system of 42 nodes was used to show the accuracy of the mathematical model as well as the efficiency of the proposed solution technique. © Brazilian Society for Automatics-SBA 2013.