A microscopic model for simulating barge traffic through a series Baumol's model. Nason and Kullman (8) developed a total of locks has been developed and tested with data for a section logistics cost model to predict inland diversions from waterof the Ohio River. The model was designed primarily to analyze ways. the economic effects of waterway congestion and service relia- Two models based on queueing theory have oeen found for bilitv. The results indicate that the model is capable of simulating the system performance sufficiently well for analytic purposes. estimating lock delays. DeSalvo (9) models lock operation as The results also indicate to what extent coaL stockouts would a simple singLe-server queueing process with Poisson distnbincrease at a power plant, or alternatively, how safety stocks uted arrivals and exponentially distributed service times (ie. would have to be increased, as traffic volumes approach capacity. MIM/I queues). Wilson's model (10) extends DeSalvo's by treating Lhe service processes as general distributions (M/G/I