Home for the Holidays

By Louise Totten and Noha Tohamy

In an industry with almost 100 percent driver turnover a year, truckload motor carriers must go the extra mile to keep drivers on board. Getting drivers home during the holidays is considered crucial to retention. At the same time, the carrier must meet the pick up and delivery requirements of its customers. More and more carriers are turning to optimization techniques to solve the dilemma.

Imagine this: It's Dec. 22, you've worked two weeks straight, usually 10 hours a day, often more, always alone. Each night you've eaten in a diner and slept in a small bunk hundreds of miles from home. Now, when you should be on your way home for Christmas, your boss says, "I need you to work two more days." Your family expects you at home, yet you don't know if you'll get there by Christmas Eve.

Almost every long-haul truckload driver faces this fear. All year, these men and women drive for two weeks at a time, getting paid for each mile run, picking up and delivering loads, anywhere in the United States or Canada. Some days the driver gets lucrative cross-country loads. Other days the driver waits, unpaid, at a truck stop until his dispatcher finds him a load. But now it is Christmas, and all that driver wants is to get home.

Most companies compensate drivers with about two days off for each two weeks on the road. When a driver is due home, the company tries to find freight that gets the driver home. However, such freight isn't always available. Then the dispatcher tries to find a load in the general direction of the driver's home, or gives the driver short loads until a "get-home" load is found. In an industry with almost 100 percent driver turnover a year, truckload motor carriers take their promise of regular home time very seriously. This promise becomes an iron-clad commitment during the holidays.

Of course, getting drivers home is not the only goal. At the same time, the carrier must meet the pick up and delivery requirements of their customers. Shippers want freight moved off their docks at certain times, and they want it to arrive at the receiver at specific times. Even if a load that would get a driver to his home town is available, the driver in question might not be finished with his current assignment in time to pick it up. The freight planners at trucking companies juggle these issues every day, trying to balance the drivers' needs and the customers' requirements.

The freight planners have another important factor to consider - the company's bottom line. The load that gets the driver home may pick up 200 miles from the location where he delivers his current load. When a truckload company tells a driver to "deadhead" empty from one location to the next, they have to pay the driver for those miles. But no one is paying the trucking company. So minimizing empty miles is always a priority. Freight planners frequently are forced to meet one need - the driver's, the customer's, or the carrier's - at the expense of another.

Now consider the holidays. Most trucking companies promise that every driver can be home for Christmas. Sometimes, carriers buy their drivers bus tickets, rent vans, and even purchase last-minute plane tickets to get the drivers home by Dec. 25. In a large truckload company, the freight planners are trying to get more than 1,000 drivers to their requested locations on or about Dec. 24, while picking up and delivering all their freight on time. How on earth do they do it?

The Two-Step Solution
Advances in operations research techniques and computing power have revolutionized freight planning. Carriers divide the country into separate regions, assigning a freight planner to each one. Traditionally, the planners have manually found the best driver-to-load pairing within the region. Today, more carriers are relying on global optimization systems that generate the best possible system-wide matching between drivers and loads. One such system is MICROMAP, used by nine truckload carriers, including some of the largest carriers in the country. Global optimization systems introduce new options to planners that were almost impossible to find using the traditional methods.

A global optimization system enables the freight planners to "herd" the drivers in the right direction prior to their target get-home date, first positioning the driver in an area that usually has outbound freight toward his home, then sending the driver on a load that delivers close to, or passes nearby, his home.

Suppose it is Tuesday, Nov. 21. Driver Suzanne is in Eau Claire, Wis., and wants to be home for Thanksgiving. Her home is in Albany, N.Y. Suzanne wants to drive as many miles as possible before the holidays to earn as much money as possible. To send her home immediately would ensure that she gets there for Thanksgiving, but would sacrifice a day of utilization, financially penalizing both Suzanne and her company. The global optimization system suggests a better decision, assigning Suzanne a load to Knoxville, Tenn., where it is likely there will be a high number of loads that she could deliver to Albany in time for Thanksgiving.

As soon as Suzanne is on her way to Knoxville, the optimization system starts looking for opportunities to get her a load that will deliver in Albany. The system helps the freight planner see beyond the Knoxville city limits to find the right load.

The carrier "teaches" its optimization system the relative priorities of the corporate objectives, which can change throughout the year. Around the holidays, the driver get-home request is frequently top priority. This contrasts with the rest of the year when customer requirements come first. The carrier's optimization system must satisfy changing priorities by being flexible and easily maintainable.

Because load and driver information is updated 24 hours a day, real-time communication using mobile satellite transceivers has become common in trucks. The optimization system needs to react to the constantly changing situation, refreshing its recommendations on a minute-by-minute basis. Suppose driver Randy's in-laws decide to come a day earlier for Christmas, so Randy needs to go home earlier than planned. Randy alerts his dispatcher using the real-time communication system installed in his truck. Within a minute, the optimization system knows of the new request and finds the best global matching given this new information.

Helping Drivers with a Bad Load
No matter how well carriers plan, they are sometimes forced to assign a driver to a load that will not take him home on time. After such an assignment is made, the freight planner looks for alternatives to get the driver home. Sometimes two drivers can meet at a truck stop to swap loads. Such trades can help one, or both, drivers get home. They simply unhook their loaded trailers and swap them.

For example, assume that on Dec. 23 driver Joe is hauling a load from Iowa to Georgia. He lives in Boston and wants to be home for Christmas. When he was dispatched, there was no freight available from Iowa to Boston. Similarly, suppose another driver, Sam, wants to go to Georgia and is hauling a load from Louisiana to Connecticut (Figure 1: Swapping loads in Tennessee helps two drivers get home). Sam and Joe could meet in Tennessee and switch loads. Then Sam could deliver Joe's load to Connecticut and Joe could deliver Sam's to Georgia, and both drivers would be home by Christmas Eve.

While swapping loads can inexpensively solve many problems, without decision support most of these chances to swap are missed, particularly if the carrier has hundreds or thousands of drivers. A mathematical modeling program called DROP & Swap can spot opportunities like the swap between Joe and Sam, by analyzing real-time information on both drivers' locations and their respective routes from a satellite communications system. The program can also be customized to achieve the most important holiday objective: sending drivers home.

Imagine if you were a driver and you ended up eating Christmas dinner at a truck stop. Would you want to keep working for the carrier who failed to get you home? Probably not. Improving a carrier's ability to get drivers home throughout the year, and especially during the holidays, has a significant impact on driver turnover.

And driver turnover has a significant impact on every carrier's profitability and growth potential. The expenses associated with driver turnover are shocking. The cost to recruit and hire a new driver averages between $3,000 and $5,000, and with the average turnover rate running 80 percent per year, the expense is enormous [1]. Some carriers spend 5 percent of their operating expenses on recruiting, training and retention [2].

High driver turnover also curtails growth. Carriers have less trouble finding freight than they do keeping drivers in their trucks [3]. The industry has plenty of room to grow. But the truckload industry will need 300,000 new drivers per year for the next ten years to satisfy the increasing demand for freight service [4]. So if a carrier can find a way to retain current drivers and simultaneously hire new ones, chances are high they will be able to capitalize on the opportunities in the market. Optimization systems to help get drivers home help make this possible.

"We are on the leading edge of technology," M.S. Carriers, a 2,500-truck carrier, states in its 1994 annual report. "With our (optimization) systems in place, we are cutting costs and improving efficiency... (The) computer systems allow us to route our drivers more efficiently, improve our on-time pick up and delivery rates, and keep our customers happier."

The 1993 J. B. Hunt Annual Report states that, "Operations management took a giant leap forward in 1993. A new software program ... was introduced to simplify the logistics manager's job by sorting through the difficult process of matching loads to drivers. This enables the computer to consid ... er over 90 different assignment factors - far more than the human brain can comprehend at one time. Already the technology is responsible for a reduction in empty miles of more than 10 percent, has contributed to getting drivers home on time, and has assisted in the trailer trade-in program."

In the same annual report, J. B. Hunt reported a reduction in driver turnover from 105 percent to 94 percent. One of their logistics managers said, "The system found a load in a neighboring region to get one of my drivers home ... I would never have seen it, and it was only five miles from where the driver was sitting." [5].

To view a graphical display illustrating the manual and direct support system (DSS) methods of optimizing a trucker's holiday schedule, click here.

The power of an optimization system to keep drivers happy is most obvious during the holiday season. The value of getting long-haul drivers home on-time drops directly to the carrier's bottom line.

"Optimization" and "operations research" have become household words to motor carriers. Not only do carriers use optimization products, but some of them drive the development of new optimization systems. For example, many carriers believe that the next contribution optimization techniques will make to motor carriers is cutting their fuel costs by suggesting better routing and fuel stop decisions.

Click here to view two sidebar stories: Switch, Swap On the Fly and Gaining Control of LTL Load-Planning Problems
  1. Richmond Times Dispatch, July 9, 1995, p. E6.
  2. "Turnover, Over and Over," Fleet Owner, September 1994, p. 18.
  3. "Chronic Shortage of Truckload Drivers Spurs Innovative, Long-Range Solutions," Traffic World, March 27, 1995, p. 30.
  4. "Seeking a Solution: TL Carriers Organize Task Force," Transport Topics, Jan. 30, 1995.
  5. "Optimization Makes J.B. Hunt More Efficient," Transport Topics, May 17, 1993.

Louise Totten and Noha Tohamy are operations research analysts for PTCG, Inc., a software development company based in Burlington, Mass. PTCG develops and sells a variety of decision-support software systems, including MICROMAP and DROP & SWAP, for the motor carrier industry that rely on optimization and other operations research techniques.
OR/MS Today copyright © 1995 by the Institute for Operations Research and the Management Sciences. All rights reserved.
Click here to return to the table of contents.