Transportation Simulation: Who Needs It?

If you have a lawn mower, why would you need a lawn service? If you have an iPod nano and a cell phone, why would you also need an iPhone? If you have a transportation management system (TMS) and a network design tool, why would you need a transportation simulation solution?

A network design tool works in strategic time frames, often a year or more, and answers questions about where manufacturing plants and warehouses should be located, and where particular products should be sourced, made and stored.

TMS does the tactical planning – a time frame of days or weeks – and is the execution tool.

Transportation simulation might be described as an intermediate tool. The time horizon covered by its analysis is shorter than a design tool and longer than a TMS. Basically, these tools take months of actual historical shipment data (forecast data can also be used), loads massive amount of data into the tool (this is truly a Big Data application), and answers the question, “What would have happened if we had done this instead of what we actually did?” The simulation shows what life would have looked like one day at a time. This is a granular view of how busy the DC would be, when transportation service would have deteriorated, and when out-of-stock situations would have occurred.

In recent weeks, I’ve heard three answers to the question of why a shipper needs an intermediate-level transportation solution. At JDC Focus, I heard a presentation on how one of JDA’s customers was using its Transportation Modeler solution (see “Vignettes from JDA Focus 2013”). Toby Brzoznewski, Executive Vice President at LLamaSoft, visited us recently and one of the things we discussed was their transportation simulation tool, Transportation Guru. And I had a detailed conversation on this topic with Chris Johnson, VP of Research & Development at LeanLogistics. LeanLogistics uses its LeanOpt3 (cubed) technology to offer a managed service in this area. Almost all of the analysis that follows is based on my conversation with Chris.

One thing a transportation simulation tool can do is help to validate the output of a network design. Network design solutions, in order to solve the complex problems they address in a timely manner, make simplifying assumptions based upon aggregated data.

But the main thing I wish to focus on is how a transportation modeling tool provides value to the transportation department specifically.

In many transportation departments, simplifying transportation policies have been developed without being tested.  Examples include: anytime we can ship intermodal, we will; anytime we can combine two less that truckload shipments to create a 25,000 pound truckload shipment, we will; anytime we can combine less than truckload shipments into a truckload multi-stop delivery, we will. These rules of thumb often go unchallenged. In theory, if the TMS does optimization on a daily basis, the system would come up with an optimized plan that is not always based on these simplifying assumptions. However, Chris pointed out that many TMS solutions have been configured to mirror existing policies and thus do not perform unconstrained optimization. By using historical data to simulate what could have been, LeanLogistics has at times found savings of as much as 3 percent of the freight spend for shippers willing to reconfigure their TMS and remove the artificial constraints.

Secondly, the transportation modeling tools can be used to see what would occur if a company could get their customers to agree to a longer delivery window. For example, if some subset of customers would agree that deliveries within 48 hours would be fine, rather than the current 24 hour delivery window, how much money could be saved? Once the analysis is done, the shipper can go to their customers, show them the savings and say, “If you agree to this, we will not have to raise our prices.”

Finally, this analysis can be used to defend the transportation costs that were incurred. A CFO may not understand why just because demand was 10 percent higher than expected, freight costs went up 15 percent. The CFO may assume economies of scale and not understand that if unexpected demand occurs on particular high cost lanes, costs will not advance in lock step with demand.

In short, if you have both and iPod nano and a cell phone, who needs an iPhone? I’ve discovered I do. If a shipper has both a TMS and the output from a network design study, who would need a transportation modeling solution? A far larger number of companies than currently have access to these solutions.

(Note: JDA Software, LeanLogistics, and LLamasoft are Logistics Viewpoints sponsors).


  1. I like your point about existing policies going unchallenged. Testing is so important in any field that wants to optimize results, minimize costs etc. It’s very true that the results are often unexpected.

  2. Great points regarding the benefits of different tools to address different timelines of decision making. This can be overlooked easily, but in each of the tool set there are a unique set of assumptions and a different set of decisions being affected by the outcome.

Leave a Reply

Your email address will not be published. Required fields are marked *