Accession Number : AD0613618

Title :   USE OF THE 'EXPECTED VALUE SOLUTION' IN LINEAR PROGRAMMING UNDER UNCERTAINTY,

Corporate Author : RAND CORP SANTA MONICA CALIF

Personal Author(s) : Madansky,Albert

Report Date : 11 MAR 1960

Pagination or Media Count : 16

Abstract : The use of two methods in the one-stage stochastic linear program is discussed: (1) replacing the random elements by their expected values (the 'expected-value solution'); and (2) replacing the random elements by pessimistic estimates of their values (the 'fat' technique). The one-stage problem and the two-stage problem are described, and the relation between the 'fat' techniques used in the one-stage problem and the so-called 'slack' techniques useful in the two-stage problem is demonstrated.

Descriptors :   (*LINEAR PROGRAMMING, GAME THEORY), STOCHASTIC PROCESSES, PROBABILITY, DECISION THEORY, THEOREMS, COSTS

Distribution Statement : APPROVED FOR PUBLIC RELEASE