Graph theory is a method of describing a transportation situation in a visual form. People familiar with graph theory can take a transportation scenario and interpret it as a mathematical graph. The researcher then can analyze the graph and make conclusions about different transportation methods and routes. Using graph theory, researchers often can find better or even optimal routes for companies planning logistics.
Linear programming is not a form of computer science, it is a mathematical tool that statisticians and econometricians use to maximize or minimize certain functions. In logistics, researchers can set up functions that represent certain aspects of transportation, such as how the form of transportation affects the money or time spent. After creating such a function and imposing constraints, (such as time, labor or resource constraints) the researcher can run linear programming algorithms on the function, thereby finding the variables that maximize profit and minimize cost.
Simulation is a relatively new tool for transportation analysis. However, it is a particularly useful one. In transportation analysis, there are myriad variables that can be considered. In other forms of analyses, the problem is often simplified, as too many variables makes many statistical and econometric analyses unwieldy. Simulation allows computers to handle large amounts of variables and output the short-term or long-term logistical consequences.
There are many other forms of statistical analyses that researchers use both to predict and test hypotheses. Tools like ANOVA and t-tests allow researchers to check transportation hypotheses, such as whether one route is more efficient than other. Other tools, like regression analysis, allows researchers to make predictions about future decisions, such as if changing the form of transportation will increase profit. These forms of analyses rely on past data.