Software developer working on "crystal ball for rates"

In these turbulent times, it's been hard for distribution managers to guess where freight rates will be tomorrow—never mind a month or year from now. I bet most managers would give almost anything to have a crystal ball on their desk to help them divine the future direction of transportation pricing.

As it happens, one supply chain software provider—LeanLogistics Inc. of Holland, Mich.—is working on this right now. To be precise, the company is developing and testing an application for forecasting rates for truckload movements. The new venture builds on LeanLogistics' existing transportation-related services. In addition to providing an on-demand transportation management system (TMS) and running an online cargo-matching and freight optimization network for shippers and carriers, the company manages transportation—procuring trucking services—for 12 shippers.

Electronic rating tools are nothing new to the transportation market. TMS vendors have long offered systems that allow shippers to compare carrier rates and performance. For instance, both Next Generation Logistics Inc. of Inverness, Ill., and Distribution Solutions Inc. of Plymouth, Mass., provide transportation management systems that include rate benchmarking capabilities.

But LeanLogistics' forecasting initiative goes well beyond benchmarking rates to identify the lowest available price. According to Chris Timmer, a senior vice president of sales and marketing at the company, the new model predicts truckload rates by lane a month ahead.

LeanLogistics is not the first to attempt to forecast freight rates. Others, including several economics firms, have taken a stab at it in the past. But LeanLogistics says it has an advantage over its predecessors in that it has a huge database of actual rates to draw upon. The company says its on-demand TMS (which happens to be called On-Demand TMS) currently processes millions of shipments per year, representing more than $5 billion in annual freight spend.

To predict where rates are headed, LeanLogistics starts with a real rate—say, $1.25 a mile for a truckload shipment from Chicago to Boston. It then performs some calculations using a proprietary algorithm that factors in truckload capacity and demand for that particular shipping lane along with such variables as equipment operating costs, fuel prices, labor rates, and past shipping behavior for the region and the specific market. After crunching the numbers, the model comes up with a 30-day rate forecast. To date, Timmer says, the predicted rates have been within 5 to 6 percent of the actual rate.

At the moment, LeanLogistics is developing rate forecasts only for shipping lanes used by customers of its transportation procurement service. But the company has plans to expand both the forecasting timeline and the scope of the service.

LeanLogistics thinks it can perfect its forecasting model and predict rates as far out as a year, according to Timmer. "We've proven we can do this on a near-term basis," he says. "We think we can do this on a six- to 12-month basis."

Once it's confident it can provide reasonably accurate forecasts over an extended period, the company is considering offering this as a separate service to shippers and carriers, many of whom would surely be interested in a crystal ball for rates. But Timmer expects that some will have reservations at the outset. "People will be skeptical of this," he says, "and the only way we can ease that skepticism is with results."