This paper investigates the application of a practical and robust control strategy for the operation of a sequencing batch reactor (SBR) used for toxic wastewater treatment. The strategy sets the operational conditions of the SBR based on the on-line information collected during the previous batch. In particular, it sets the exchange volume of the reactor, as well as the batch reaction duration by optimizing the amount of mass of substrate in the influent treated per time unit. The optimization uses an experimentally calibrated mathematical model of the previous SBR cycle, found using the on-line dissolved oxygen concentration measurement data. The results show the applicability of the methodology to treat synthetic wastewater containing 4-chlorophenol as model toxic compound as sole source of carbon and energy. It correctly detects changes in the influent concentration and appropriately sets the operational parameters of the process.