This letter proposes an absolute value based method to solve the interval linear optimization problem with application to security constrained economic dispatch (SCED). To avoid expensive computation associated with solving combinatorial linear programming (LP) problems for interval upper bound, firstly a bilinear programming model is formulated using duality theory. The equality constraints are then converted to absolute value constraints. Lastly, the absolute value operator is eliminated through introducing two nonnegative slack variables and complementary slackness condition. The resulting new bilinear programming model can be effectively solved by the branch and bound method with linear relaxation technique to obtain the global optimal solution. Numerical results demonstrate the effectiveness of the proposed method in improving solution and computation time.