I have been building models of customer willingness-to-pay in the trucking industry.

Generally our models have been taking the form of "revenue per move" as the dependent variable and "travel miles" along with several other variables as the independent variables.

One of the other independent variables has been "total lane volume", which is simply the number of times we have to move back and forth in a given lane for a given customer over the course of a year. Everything else equal, I would expect that the "revenue per move" would be lower where the "total lane volume" is higher.

I have been modeling "revenue per move" as a simple linear function of "total lane volume". I have been generally getting the right +/- signs and pretty good-looking p values. My feeling is that the relationship between these two variables is not a simple linear relationship. As well, when I do a scatter plot of these two variables, I find that the effect of "travel miles" so significantly swamps that of "total lane volume" that it is hard to get a good picture of the relationship.

I would be grateful for any advice on how to model the relationship between "revenue per move" and "total lane volume". I suppose the more general form of my question might be: In multiple regression, how does one prepare scatter plots of the relationship between the dependent variable and each of the independent variables in situations where the effect of one independent variable on the dependent variable swamps that of the others?

Many thanks,

Trucker37