Reducing carbon dioxide ($CO_2$) exhaust has become a major issue for society in the last few years, especially since the initial release of the Kyoto Protocol in 1997 that strictly limited the emissions of greenhouse gas for each country. One of the primary sectors affecting the levels of atmospheric greenhouse gases is agriculture where $CO_2$ is not only consumed by plants but also produced from various types of soil and agricultural ecosystems including greenhouses. In greenhouse cultivation, $CO_2$ concentration plays an essential role in the photosynthesis process of crops. Optimum control of greenhouse $CO_2$ enrichment based on accurate monitoring of the added $CO_2$ can improve profitability through efficient crop production and reduce environmental impact, compared to traditional management practices. In this study, a sensor-based control system that could estimate the required $CO_2$ concentration considering emission from soil for cucumber greenhouses was developed and evaluated. The relative profitability index (RPI) was defined by the ratio of growth rate to supplied $CO_2$. RPI for a greenhouse controlled at lower set point of $CO_2$ concentration (500 ${mu}mol{cdot}mol^{-1}$) was greater than that of greenhouse at higher set point (800 ${mu}mol{cdot}mol^{-1}$). Evaluation tests to optimize $CO_2$ enrichment concluded that the developed control system would be applicable not only to minimize over-exhaust of $CO_2$ but also to maintain the crop profitability.