Most integrated iron and steel corporations have built on-site power plants (OSPPs) to reduce their purchased electricity and thus to decrease the overall electricity cost. Due to their large quantities and easy access, the byproduct gases generated in the steel production process are the main fuels used for the OSPPs. The introduction of time-of-use (TOU) electricity pricing in the steel industry has made it possible to decrease electricity costs through an optimal collaboration between the energy storage equipment (gasholders) and OSPPs. In this paper, a byproduct gas scheduling model based on mixed-integer linear programing (MILP) considering the TOU electricity pricing is proposed. In this model, Pareto optimality and fuzzy sets were used to find the best compromise solution for two conflicting objectives: achieving the gasholder stability and reducing the electricity purchasing cost. In addition, the influence of the operation load on the boiler efficiency was considered to improve the model accuracy. The results show that the optimisation can achieve better peak-valley shifting of the electricity generation and decrease the electricity purchasing cost by 29.7% with improved gasholder stability. Optimisation increased the overall boiler efficiency by 3.3%, indicating that the byproduct gases are effectively and efficiently used. The sensitivity analysis results indicate that the peak-valley shifting of the electricity generation improves with increasing peak-valley price rate (PVR) at the expense of decreasing the overall gasholder stability. (C) 2017 Elsevier Ltd. All rights reserved.