According to ISO standards life cycle assessment (LCA) consists of goal definition and scoping, inventory analysis, life cycle impact assessment and interpretation. In goal definition and scoping LCA system boundary is defined. Since LCA is time consuming then there is a need for a systematic approach to determine which processes needed to be included in the system boundary. This paper fulfills the need by proposing an improved binary linear programming model for LCA system boundary identification. The objective function of the model is to minimize the number of processes included in the system boundary and its constraints are the specified mass, energy and economic value ratios. In order to demonstrate its applicability an example is presented. A sensitivity analysis is also conducted in order to illustrate how the change in the specified mass, energy and economic value ratio will affect current optimum system boundary.