I thought it would be interesting to talk about specific types of simulation models and the usual objectives of each type of application.  One application that has been more popular in recent years is the simulation of inbound and outbound truck and employee traffic at automotive facilities.  

Sample Traffic Simulation Image 

There are a number of reasons to undertake a study of this type:

1. Validating the gate staffing versus the planned process at the gate
2. Determining estimated dock utilization
3. Determining the time for each sequenced commodity to get from the site to its dock
4. Evaluating employee arrival and departure times and routes to reduce the impact on the delivery of just-in-time sequence deliveries.

Gate Sizing and Staffing:

Usually a site has more than one entrance.  Specific vendors will choose the most convienent entrance unless they are told (forced) to use a specific entrance.  Simulations can be used to balance the workload at each gate and reduce the average wait time for a truck to enter or exit the site.  As part of this process, trucks are classified into sequence versus non-sequenced commodities.  The non sequenced materials can show up any time.  they are usually less frequent and the driver is more unfamiliar with the site.  The check in time for these trucks is usually longer and there may be an inspection, directions, or specific paperwork required to enter the site.

Sequenced arrivals are often given a dedicated lane and a pass card so they can enter the site without guard interaction.  These drivers are usually at the site several times a day and they are taking a dedicated route to a specific dock every time.

The simulation can also assess the impact of inspection philosophies.  Some plants do a 100% check on outbound trucks to look for contraband materials.  Other sites have a random check.  The simulation can quantify how different inspection strategies impact the average queue at a gate and the average delay time for sequenced materials arriving at the site.

Dock Utilization

Some sites will have outside drivers drop their trailers in a marshalling yard and pick up an empty trailer and leave.  The OEM is then responsibile to accomplish all the trailer moves on the site with a dedicated crew of tuggers.  Other sites let vendors drive to a dock where they unhook their load and grab a trailer from an adjacent dock.   Some sites require the vendors to wait for their parts to be unloaded and empty containers to be reloaded before they exit the site with the same trailer.  Lastly, some trailers have dedicated automation inside the trailer that allows the trailer to act as an automatic buffer for the plant.  In each of these cases, the utilization and number of docks required by the plant is different.  Simulations can quantify these choices and help determine the material handling staff inside the building required to support the arriving commodities.

Broadcast Impact of Site Traffic

Sequence parts often arrive on very tight windows.  Parts are built at a local facility and shipped to the plant just before they are needed.  As part of planning this arrival, vendors need to know how much time they have to build a component and get it to the appropriate dock.  Once a vendor knows the lead time they will have to make a part, they need to subtract the time to build the part, the time to load it on their truck and the time to drive it to the facility, and the time to unload it at the OEM.  Vendors often forget the time that they have to wait due to traffic at the gate, the time inside the vendors property to get to the dock, and any congestion on the site.   Simulations can give the vendors estimated times that they can deduct from their broadcast windows when trying to estimate the number of trucks and how many parts can be loaded on each truck.

Employee Traffic

Recently we studied a few automotive sites to assess the impact of employee arrivals on truck traffic.   The key times are obviously during shift changes.   The models are fed employee demographics by zip code to determine the most likely arrival point at the plant.  Employee arrival and departure time distributions are modeled using data collected from either the existing site or a similiar site in the same area.  Once the model shows the impact of the employee traffic, experiments are developed to assess the impact of dedicated employee arrival gates and if the local city planning office can assist by changing traffic lights or expanding exit ramps to ease congestion.  This is also a useful tool for helping vendors adjust their daily schedules to make sure more parts are available to the plant at these shift changes to avoid any shortfalls due to accidents, weather, or other congestion issues.

I am planning to write about other applications in the coming weeks.  Please suggest a topic that you would like to see discussed.