Connecting A Missing Link: Transportation Business Intelligence

By Gary Girotti, Vice President, Transportation Practice, Chainalytics LLC Monday, July 27, 2009 I have found myself drawn into a new area of transportation that deserves...

By Gary Girotti, Vice President, Transportation Practice, Chainalytics LLC
Monday, July 27, 2009

I have found myself drawn into a new area of transportation that deserves a lot more attention – business intelligence. I’m naming it BI instead of metrics because I think it needs a label with more status and importance than metrics provides. BI does that.

But it surprises me that we don’t see more focus on BI in transportation. If any business function deserves business intelligence it should be transportation management.

Although admittedly I have a biased perspective, who would question the need to increase the understanding of what drives transportation performance? For one, there continues to be a lot of mystery about transportation. More important, though, is the fact that transportation is by far the greatest cost in supply chain –totaling 61% of the $1,397B logistics market.

Today in this post I want to share a small breakthrough in our work on transportation BI. We have developed a framework from which meaningful performance drivers can be clearly isolated, which is the essence of BI – improved understanding through better use of key information. In later posts I will hopefully report on some relevant findings and successes of this new development.

To begin with, I’ve concluded the reason we don’t see more on transportation BI is because measuring transportation cost is extremely variable and complex. Networks, flow paths, and transportation cost drivers are constantly changing. With so many variables, changing it is hard to grasp what makes a difference. Remember when studying simulation, the key was to only change one variable per scenario? How so in transportation, with so many changing variables? The complexity is so great that even the better transportation management systems (TMS) offer very limited modules that require customization to get value from the metrics.

Let’s be more specific about variability and complexity. Variability means base costs can vary dramatically over short time periods. The current environment offers a good example. In June 2008 fuel was over $4.00 per gallon. In June 2009 it was about $2.40 per gallon. From a rate perspective, we remain in a depressed market, and our benchmarking studies show that rates have dropped 8% since last year. Further, in today’s market spot market rates are actually lower than many contracts.

Transportation is complex. Primary cost drivers vary by industry and by flow path. The question for any individual business is what factors make a difference in driving performance? The answer to this question is often different within the same business. For example, the inbound manager for a retailer will focus on different factors than the outbound manager. For some it is mode, others distance, and others still load utilization. It is difficult to develop fair comparisons because network and flow path complexities result in varied drivers of performance.

So, the real challenge in transportation BI is to find an effective way to:

  • Determine the most important variables
  • Isolate those variables

And this is where we’ve had an innovation.

My colleague Brian Fish has developed a process for determining the most important variables and a two-step BI framework to isolate these variables for analytics. The result is a set of meaningful views that enable managers to spot where costs are out of control. This clever and unique approach utilizes common tools like process flow charts and fishbone diagrams that users can understand. The isolation framework is really cool; it separates costs by lane (from which cost/mile vary tremendously) and then by individual cost drivers like load factor or number of stops. Separating the lane cost from the others is the critical part of this solution. As you well know, the cost of trucking per mile varies dramatically depending on lane. For example going 500 miles into Florida is incredibly expensive vs. coming out 500 miles. Isolating lane costs first takes away a key market variable, enables a much more meaningful view of the other cost drivers.

We are really encouraged about this new innovation and are in the beginning stages of implementing client results from this process. Stay tuned, as we have more to show as we draw on our findings.

In this article