One solution to the problem of increasing demands on our water resources is optimal allocation of available water. In this study, a linear programing model was used for optimizing water productivity for major crops grown in the Geray irrigation scheme. The model objectives were benefit maximizations subject to land and water availability constraints. The linear programing models were solved by using Micro Soft Excel Solver and developed under optimal water supply and three different levels of deficit irrigation (10%, 20% and 30%) and were compared with the existing scenario in terms of crop yield, total profit and water productivity. In the existing irrigation condition (scenario I), the area covered under cultivation was 415 ha and the yields obtained were 4961.6 tons, the net benefit was 62.18 million birr and also water productivity was 15.49 ETB/m3. In scenario II, the model found that, increase irrigation area by 148.9%, yield by 155.6%, benefit by 155.7% and water productivity by 101.3%. By comparing the two scenarios, changing cropping pattern and applying full irrigation practice (scenario II) was more advantageous than scenario I. And also, the optimal cropping pattern developed under irrigation deficiency of 10%, 20% and 30% (Scenario III, IV and V) increased yield by 140%, 127.9% and117.5% respectively, net benefit by 136%, 120% and 106.6% respectively, and decreases water productivity by 1.88%, 2.5 % and 1.23% respectively and also increased total irrigated land by 148.9% as compared to Scenario I. Using irrigation deficiency of 10%, 20% and 30 % (Scenario III, IV and V) the optimal solution of the model decreases yield, net benefit and water productivity as compared to scenario II. Finally optimal cropping pattern developed under full CWR (scenario II), gave the maximum total profit and optimum water productivity compared to scenario I and scenario III, IV and V. Therefore the actual irrigated land during the study period was below the optimal values, and water released for irrigation was poorly managed. Thus with optimal crop planning and water management practice, it is possible to irrigate the designed command area. Generally the current irrigation water management practice in the scheme was poor; therefore it needs enhancement to get maximum profit and maximum yield. And also to make the project profitable, it is preferable to use high-value crop areas, with small area which have low water requirements. Key words: optimization, linear programing model, water productivity, cropping pattern