The simulation model
The data set used in the simulation is based on a typical four-week flow through the terminal of a national LTL carrier. The format of the raw data provided by the carrier
is tabulated in Table I.
The data provided associated with the number of truck arrivals fit the Poisson distribution with mean 5/h. The inter arrival time between each truck is determined to
be exponential with mean 12 min. A truck is assigned an origin with a certain
probability based on the inbound freight volume and directed to the appropriate strip door. The number of pallets in each truck is determined based on the uniform distribution with minimum 25 and maximum 50. This assumption is founded on the fact that a trailer can accommodate approximately 25 single-stacked or 50 double-stacked pallets. Each pallet has a destination which is determined by a probability based on the outbound freight volume. To simulate the movements of the pallets within the facility, the actual dimensions of the facility are used. The section of the facility where pallets are transferred is a rectangle with length 264 in. and width
84 in. and each dock is 12 ft apart from each other (center to center). The layout of the facility with existing dock assignments is shown in Figure 2. This information is used
to calculate the rectilinear distances between every pair of docks. To determine the time taken to transport the pallets between doors, the speed of the forklift is assumed to be 50 ft/min. It is also assumed that the loading and unloading times of a forklift follow
a uniform distribution with minimum 0.3 and maximum 0.4 min. The total number of truck departures is assumed to be equal to the total number of trucks arriving, and
their destination and inter-departure times are based on the outbound freight volume. The inputs to the simulation model are summarized in Table II.