The recent literature on power system reliability has emphasized the importance of sound planning to satisfy future loads. In view of the extremely high investment costs of electric power systems, it is imperative to have procedures for adding the right kind of equipment at the right time in the right location to achieve the desired level of reliability and quality of service at an acceptable cost. These procedures are very complex because of the large number of design alternatives that need to be reviewed from the points of view of economy and technical performance (including reliability). In this report, power system planning is viewed as a problem of mathematical optimization in which the sum of long-term investment and operating costs is minimized subject to reliability constraints. This mathematical optimization provides the desired procedures for system design and gives the most economical schedule for system expansion. Digital computers have been used extensively during the past decade for simulating the expansion of a power system. Direct optimization, however, constitutes a new development; its feasibility is proven and illustrated in this report. In addition to outlining new methods for direct optimization, this report develops efficient methods for computing the operating and capital costsmore » and for assessing the reliability properties (including transient stability) associated with each design alternative. Finally, the problem of expanding a power system in the presence of uncertainty about future load, technology, financing costs, etc., is viewed as one of stochastic optimization, and procedures for computing the optimum decision policy are developed and illustrated.« less