Refining The Process Of Short-term Production Planning Decision Support System

Oil refinery industry is one of the typical representatives of process industry and plays an essential role in national economy. One of the important issues for oil refinery is comprehensive automation. Researches on this topic have been conducted for the last several decades and a lot of results have been received. However, up to now, it lacks efficient techniques and tools for executive system (MES) at the optimize layer, one of the cores of comprehensive automation. While there are efficient tools for long run scheduling using linear programming, short-term scheduling is still done manually. This greatly obstructs the advance of informalization for oil refinery enterprises. Thus, it is significant to develop an efficient software tool to help planners for short-term scheduling.With a real oil refinery plant as a case, investigation for the characteristics of oil refining production process and the status of research and application in short-term scheduling has been done in detail. It is found that there are a lot of constraints and it involves very time consuming computation in short-term scheduling. Thus, we conclude that a "heuristic + simulation +optimization" may be the realizable way to develop such a software tool to overcome the difficulty faced. Based on this idea combined with DSS concept and some optimization rules, in this paper we develop a computer aided decision support system for short-term scheduling.In this thesis, we first present the requirement analysis for the system, the methodology in designing the system, as well as the workflow in the system in detail. Then, two key models used in the system are given: 1) the model for production state simulation; and 2) the decision aided model. Examples are provided to show the application of the models. Finally, we show how the system is implemented.