Quasi Random Sampling for Operations Management

Abstract

We look at the benefits of using a kind of quasi-random numbers to

obtain more accurate results for a given number of simulation runs. We

explore a sampling method with enhanced independence in multidimensional

simulations by combining the ideas of stratified sampling

and Latin Hypercube sampling. We test the new sampling method by

comparing it with traditional stratified sampling and Latin Hypercube

sampling applied to various operations management problems.