MIT's Caplice and the Three Myths of Transportation

One of my favorite supply chain academics is Dr. Chris Caplice, executive director of MIT'S Center for Transportation and Logistics, because he leads research that is in the end highly practical and usually quite relevant for real supply chain professionals. He also adds to that an ability to articulate that research very effectively - both qualities not so common in academia.

So we were very pleased to have him as the keynote speaker during our recent Transportunties on-line conference and exhibit, and I had a good time doing a video interview with him for that event on "Three Myths of Transportation Management."

It was very good, so I am going to take the easy route this week and summarize Caplice's presentation.

Number 1: The Myth of the Market Rate: This one frankly surprised me a bit. Of course, there is great interest by shippers in identifying a "market rate" for a given lane, because naturally enough they want to know what they should be spending for that move. In great summary, Caplice says shippers are often largely wasting their time.

My first question to Caplice was whether market rate means an average type rate or the best rate, and he said that can be a little fuzzy depending on the company. In general, however, while many would like to identify the best rate, most firms would be happy to just know the average. Many sign up to various benchmarking services out there to try to glean that intelligence, or do research work on their own.

The first challenge, Caplice said, is that it is actually very hard to find enough data. Caplice and some others received full transportation details from 100 companies, and across all those shippers, amazingly, only 3% of lanes had data from more than two companies at the 5-digit zip code level. It was better at the 3-digit zip level - 25% had more than two shippers - but still the matches in general were too few to really draw strong conclusions.

Confidentiality agreements with carriers serve as another barrier, Caplice says, but the real challenge is simply that "the market rate doesn't exist." To prove that point, he showed analysis from one lane for which there were about 10 shippers participating, with a total across them of some 3000 moves a year.

As you can see in this chart below, across a full year the average rate across all these shippers varied dramatically. While the full year the average was $2.93 per mile for the 450-mile haul, during the year the average at any point in time ranged from a high of about $3.03 in February to a low of less than $2.80 in September. So what is the market rate?

Caplice showed it gets worse if you plot rates over this period by individual shipper. Some are consistent across the year in rate paid, some vary wildly up and down, some suddenly have their rate drop substantially (a new bid event?), etc.

"Every single lane looks like this," Caplice said. "It almost looks like a blood splatter."

Truckload transportation is simply not a pure commodity as is often thought, he added, with every shipper having different needs and requirements, while the market backdrop is also changing a lot through the year.

The bottom line: don't spend too much time, energy or money trying to identify a market rate for a lane when that just doesn't really exist. Benchmarks may provide some intelligence at times, but really should serve as just as one input into a complex equation as to what any given shipper should be paying.

Number 2: The Myth of Combinatorial Auctions: This was a hard one for Caplice - he had actually done his doctoral dissertation on the opportunities from this approach, and touted the potential through much of the first decade of this century.

What does this involve? The theory is that if carriers can bid simultaneously on a bundle of lanes, they can offer a lower total cost because of the resulting network and asset efficiencies a given bundle would generate for them.

The problem, Caplice said, is that while the benefits are there in theory, in practice the concept falls apart, for a variety of reasons.

First, few carriers actually submit such combination lane bids - less than 25% of carriers in fact to those shippers using this combo approach. Even those carriers which do submit combo bids usually do so for just a few lanes, or even just one combo. Second, 50% or so of the combo bids involve a simple out and back - to a destination and then back to origin. That is much more basic and simpler bundle than the combo method envisioned. While there were a few bids that involve multiple moves in a sort of tour that would end back at the origin and other more exotic moves, these were in fact very rare.

Then there is the practicality of even these simple roundtrip moves. The carrier and shipper are betting that these out and backs will be there on a timely enough basis to make it work sometime in the future on a consistent basis. Anyone want to wager on that one? Now try making that work in even more complex combos. Not going to happen.

On top of all that, these combos are rarely even less costly than just bidding out individual lanes. Almost unbelievably, in the data set Caplice used, the combo bid was only less expensive that the individual lane bids from the same carrier 16% of the time. We can thus infer that carriers presenting combo bids rarely win them.

Logically enough then, in the rare cases when a carrier does win a combo bid, shippers rarely tender a set of loads in a way that matches that combination.

To be clear, there is value in doing carrier bid optimization using fancy tools at the lane level, but extending that to combo bids, as should now be clear. is unlikely to deliver any real value.

I will note as an aside that all this is also what is behind the great challenge shippers have in doing collaborative shipping with others, another concept that makes great sense in theory but is devilishly hard to get to work in practice.

Number 3: The Myth of Optimized Transportation Plans: Ok, this isn't so much a myth as a set of questions and observations, which I will summarize like this: Is it better to build in "robustness" into a supply chain and logistics function, to minimize the chances of disruptions, or to go to a leaner (and lower cost) model and be very good at reacting to problems when they arise? Planner versus responder.

What a great question, which is at the heart of risk management, isn't it? Yet, how many companies really analyze and quantify it along this dimension? So, as an easy example, you could build a large dedicated or private fleet that insulates you from most capacity type issues that at various times plague others using all common carriers, but likely at higher total cost.

MIT did a project for Walmart a few years ago trying to optimize that balance between robustness and cost, and Caplice implied many other shippers have similar opportunities. The "myth" aspect of this is that many shippers probably think they are getting this right today, without really having thought about it much at all.

MIT is working on models that will provide guidelines for shippers relative to managing this tradeoff - not sure when we will see it, but it looks like very interesting and valuable work.

I am out of space. Really good stuff from Caplice. The Transportunities show is still up on-demand, and you can watch Caplice's keynote here.

What is your reaction to Caplice's three myths? Agree or disagree? Let us know your thoughts at the Feedback button (email) or section (web form) below.

New On Demand Videocast:

The Impact of Vendor Lead Times and Variability on Inventory Levels

Introducing the New Lead Time/Inventory Level Calculator!


Featuring Mark Krupnik of Retalon, Kevin Harris of Compliance Networks and SCDigest editor Dan Gilmore.

On Demand Videocast:

The Internet of Things: Where it's Headed, and Practical Strategies for Leveraging RFID Right Now

We are Entering a New Era of "Turbo Visibility" - Learn How Connected Assets and Data-Driven Analysis are and will Make Companies More Efficient

Featuring Tom O'Boyle, Director of RFID, Barcoding, Inc., McLeod Williamson of Zebra Technologies and SCDigest editor Dan Gilmore.

Now Available On Demand

On Demand Videocast:

Achieving Supply Chain "ESP" with Advanced Simulation Techniques

Enterprise Simulation Planning Enables Companies to See the Future of Their Supply Chain - Learn How

Featuring Toby Brzoznowski, executive vice president at LLamasoft and globally recognized authority on supply chain network design and SCDigest editor Dan Gilmore.