A novel short-term planning model of the ethylene plant that incorporates the operating variables and energy utilization in both the thermal cracking and the down-stream process is proposed to explore the potential for increasing the production margin and reducing the energy losses. A multi-period mixed integer nonlinear programming (MINLP) model is formulated to attain the scheduling of parallel furnaces and the energy distribution of the overall plant, along with the determination of the key process operation involving the coil outlet temperature (COT) and coke deposition. The behavior of the product yields and coke formation in terms of varying COT profiles is investigated to enhance the profitability of the whole plant. A real industrial example is investigated to exploit the performance of the proposed model. The results show that the integrated approach attains an improvement in overall profit and achieves significant enhancement in energy savings, compared with the original optimization approach. (C) 2016 Elsevier Ltd. All rights reserved.