Capital Budgeting with Mixed Integer Linear Programming: An Application

Abstract

Where capital rationing exists among on-going and proposed businesses, a firm cannot rank independent investments using any of the conventional evaluation methods [1]. The problem arises because of the thousands of individual investment opportunities, only a small fraction of which can be identified over a twenty-year planning period. The standard financial decision models are used in short-range tactical planning when future funds have been allocated and identifiable projects are being considered. For long-range planning, a linear programming approach can be used to find the level of investment in these businesses that will maximize the profitability of the company through time. Many institutions have adopted such corporate models [5, 7]. These models take many forms, but they are all used to generate more timely and accurate economic planning information. This paper describes the structure of a linear programming model and how it is used in the strategic planning process by Cities Service Company. The structure of the model used by Cities Service Company was tailored to the needs of corporate planning and the nature of information available. Corporate planning required a model that would depict levels of activity, risks, and opportunities in the company's diverse businesses according to the varying (and possibly conflicting) objectives of the corporation. This model was created to 1) select the best alternatives consistent with the objectives of the corporation and the available resources; 2) aggregate the forecasts and project corporate financial statistics; 3) determine the sensitivity of the selected alternatives to varying objectives and constraints; and 4) show cash bottlenecks for possible rescheduling of capital expenditures, projection of borrowing requirements, and anticipation of dividend capabilities.