TCA Summer Profitability Seminar


On Wednesday, I’ll be presenting at the first of the TCA’s Summer Profitability Seminars in Chicago, this one with a focus on taking control of and effectively using data. This will be an excellent opportunity for me to network, hear from carriers about what pain points they’re experiencing in their own operations, and talk about some of the cool stuff that we’re doing with data at AM Transport. In this new age of big data, it’s critical that market participants use the tools available to them to stay ahead of the curve and drive efficiency gains anywhere they can be found.

Many of our readers will be familiar with the concept of the Internet of things (IoT), the phenomenon by which more and more objects are gaining internet connectivity and electronic hardware which are used to both communicate and interact with the wider world. Many in our industry may already have access to a mass of data being delivered to them via IoT devices, the most obvious of which are the newly-mandated electronic logging devices; sometimes the sheer amount of data available is overwhelming, so the question is: how do we use these data in meaningful and impactful ways to improve our own operations?


At AM Transport, we’ve implemented some “home brewed” strategies that we’ve seen good success with: our rate calculator, AI-generated market intelligence, and market forecasting in the form of a map displaying short term anticipated rate shifts.

Above we can see a historical example of our STARS Map. This takes inputs from a variety of sources and is populated every morning for our team to visualize where we anticipate rate shifts will occur and the potential magnitude of those shifts. 


We’ve also developed a rating tool that we’re quite proud of and have seen great success with. Our team simply has to enter the zip code of a real or hypothetical lane, and the model will run behind the scenes to return both a proprietary rate range as well as data from DAT and a paragraph of AI-generated text explaining, in plain English, the market conditions in that lane. We learned very quickly that while computers do well with interpreting a range of numerical data, humans are terrible at it, and so we developed this tool to return market intelligence that is easily digestible for our team.

This machine learning model takes a variety of near and real-time inputs and has the inherent advantage of delivering a rate based on what is currently happening in markets rather than looking back in time at what has happened.


It’s important to remember that data and data analytics isn’t a silver bullet that will instantly solve all of our problems, and some degree of skepticism around the implementation of anything new is healthy. Get eyes on any new technologies implemented from what I like to call your “tribal knowledge elders” to verify their efficacy, but recognize that insights from data are a powerful tool which, because of massive reductions in the cost of processing power, is within our reach.