Linear Programming A Problem-Solving Tool for Petroleum Industry Management

Abstract

Abstract The areas are identified where the techniques of operations research have been applied in the petroleum industry. Particular emphasis is given to the linear program models that have been used for solving management problems. Of the entire gamut of applications for linear programming, attention is focused on the models that represent underground reservoirs. Such models could be used to help schedule drilling-rig operations and to schedule oil production. A framework is established for placing the model for underground reservoirs in the proper perspective along with the models which represent other facets of the oil business. Introduction One of the remarkable phenomena of this century is the rapid growth in the size and complexity of human organizations. The mere size of modern business means that administrative decisions can affect vast quantities of capital and a large number of people. Mistakes can be tremendously costly, and a single wrong decision can require years for rectification. Moreover, the pace of modern business is such that decisions are required more rapidly than ever before; the mere postponement of action can give a decided advantage to a competitor. It is not surprising that the increase in difficulty of decision-making has been followed by efforts to put this activity on a more objective basis. The growing science of operations research, which relies heavily on the methods that have proved so successful in the physical sciences, is one aspect of these efforts. There is every reason to believe that the petroleum industry will reap many benefits from the good work that is in progress which every day proves over and over again the worth of operations research. The petroleum industry has long benefited from research on products, processes, geophysics and crude production. However, only recently has it applied the scientific method to solving management's problems, making full use of the methods of operations research coup]ed with the use of computers. No attempt will be made in this paper to give a comprehensive definition of the term "operations research". Reference, instead, is given to two books on the subject. A direct quotation is given from a paper that deals with operations research in oil marketing and, therefore, serves as an illustration."Operations research realistically tackles complex problems in marketing and distribution and related areas of business using many powerful techniques borrowed from the physical and behavioral sciences and advanced mathematics and statistics. It often employs electronic computers for high-speed data processing to better simulate and predict the probable results of all sorts of actions or decisions".Any company, large or small, can profitably make use of the techniques of operations research. A team usually is established to work on a problem; this should include not only scientific business researchers, but also highly experienced people from the business operations under study. This insures that pertinent points of view are taken into account in appraising the operations. The first objective of all operations research study is to define the problem and devise what is known as a model of the operation. The model, in the operations research sense, is seldom mechanical. As a rule, it uses the shorthand of mathematics to portray an actual physical operation or a set of operating conditions. Linear Programming The main emphasis in this paper will be on linear programming. The techniques of linear programming have made many inroads into the petroleum industry. Indeed, the linear program activities are so broad today that it is difficult to review and explain them. Therefore, no attempt will be made to present a complete review of all linear program projects that are active in the petroleum industry. A few major areas include the following:blending gasolines;refinery models;allocation of transportation facilities for shipping products from refineries to terminals;integrated operations-simultaneously consider the complete refinery models (Item 2) along with the transportation models (Item 3). JPT P. 729^