In chemical industries, waste solvents are considered to be the main source of pollution, whether it involves a batch process or a continuous process. Thus the separation of in-process solvents (EBS) from waste solvent streams, which is considered as process design, and the selection of environmentally benign solvent (EBS) to retrieve IPS, which is considered as product design, and are two important issues in waste solvents treatment problems. Because these two procedures could interact, it is necessary to integrate them into one framework based on the concern of cost, environmental quality, etc., to ensure better performance. In this work, we present an approach that utilizes three different methodologies at three different steps. In the process design step, the combination of residual curve maps (RCMs) analysis and P-graph technique are used to identify a separation superstructure. In the product design step, the computer-aided molecular design (CAMD) results obtained from previous work (Mm, K.-J.; Diwekar, U. M. Ind. Eng. Chem. Res. 2002,41, 1285-1296) are used as replacements for conventional solvents such as ethyl acetate. A multi-objective optimization (MOP) framework under uncertainty, in which process design and product design are combined together and with solvents recovery rate, process operability, and environmental impacts like LC50, LD50 included in the objective function, is developed in the Aspen Plus simulator. The methodology is presented in the context of continuous separation of the acetic acid-water system. More Pareto optimal solutions have been identified in this work compared to the earlier work of Kim and Diwekar.