Does load optimization live up to its promise?

In theory, it sounds like a sure winner. But in practice, it's not quite that simple.It sounds too good to be true. Let a computer program make your load planning decisions and you'll lower deadhead miles, improve on-time performance, make your drivers happier, and ultimately reap increased profitability. If you're a less-than-truckload carrier, the same type of computer magic will balance traffic

In theory, it sounds like a sure winner. But in practice, it's not quite that simple.

It sounds too good to be true. Let a computer program make your load planning decisions and you'll lower deadhead miles, improve on-time performance, make your drivers happier, and ultimately reap increased profitability. If you're a less-than-truckload carrier, the same type of computer magic will balance traffic lanes, decrease transit times, and improve asset utilization with the same result -- increased profitability.

But it is true. Load optimization technology can deliver on all of those promises. However, it can also be costly to acquire and then can consume huge amounts of managers' and technicians' time to get it up and running. It can incite uneasiness, if not outright rebellion, among valuable and experienced operations staff. And it can end up as an expensive, unused example of misunderstood and misapplied technology.

So why have some fleets enthusiastically embraced load planning and optimization tools as the key to their future growth and success while others see little value in the technology?

The answer starts with some simple definitions. Load optimization is part of a broad category of software tools known as decision support systems, which are computer programs that sift through vast amounts of data and suggest possible solutions to complex problems. One good example of a decision support system widely used by trucking is routing software, which can consider all the possible roads connecting two or more points and suggest an optimal route based on criteria such as time, distance, traffic conditions, driver preferences, and so on.

Load optimization is a far more sophisticated tool that considers routing as just one of many variables. Rather than concentrating on a single move, it attempts to deal with all the freight moved by a fleet. Most people divide optimization systems into strategic and operational tools, with strategic programs looking at long-term fleet requirements and operational systems dealing with immediate decisions. While both truckload and LTL carriers use strategic and operational optimization systems, the truckload and LTL tools perform very different functions.

In a truckload fleet, a strategic system looks at historical data, usually for the previous month or week, and pinpoints profitable and unprofitable portions of the fleet's business. It may also forecast equipment requirements for a week or more into the future to help a fleet make decisions about asset utilization.

Operational systems help TL planners and dispatchers make minute-by-minute load matching decisions, suggesting the best load for a particular driver based on a wide range of variables including customer service, driver preferences, and backhaul potential. Stepping up one level in sophistication, some operational systems can also optimize a truckload carrier's asset utilization by suggesting trailer swaps.

Strategic optimization tools for LTL carriers forecast overall freight flow so they can optimize linehaul moves and let terminals prepare load plans. They also help fleets devise long-term capital investment plans for both rolling stock and buildings, and then compare those long-term plans to actual business conditions.

In the LTL arena, operational systems compare actual freight flow to the monthly strategic plans, making suggestions to help fleets best handle day-to-day fluctuations. They also monitor inbound loads, routing trailers to particular dock doors to minimize cross-dock movements and determine manpower requirements to deliver that freight. Newer systems can also use wireless communications technology to help build outbound loads while P&D drivers are still on their routes.

Why it fails "Technology won't make you better or more profitable unless you're already good at what you do," says Larry Johnson, director of customer service for M.S. Carriers. "I'm a big fan of optimization (tools), but I'm also very cautious about them. It's really more about people and (business) processes than technology."

One of the country's largest truckload carriers, with approximately 3,000 tractors and 8,000 trailers, M.S. installed its first optimization program in 1992. "It's a pre-dispatch optimization system," says Johnson. "It looks globally at all your trucks and loads, and then recommends the best truck for a particular load as well as alternatives."

The system makes its suggestions based on tables or weighted preferences specified by the fleet. For example, routing a driver home might be deemed more important than backhaul potential, but less important than on-time delivery.

"We spent 11 or 12 months telling the system what to think," says Johnson. When M.S. actually turned the program on, it saw immediate positive results. "That first year, we got every driver home for Christmas under a load of freight," Johnson points out.

Two years ago, the fleet added post-dispatch optimization with a trailer swapping program. "We use it to get a delayed load back on schedule, to meet driver requests, to create capacity in areas where we don't traditionally have trucks, and to use regional drivers to move longhaul freight," says Johnson.

The results were impressive, he says, and it made him rethink the way the fleet used its pre-dispatch system. "Now I can separate asset utilization functions from driver satisfaction functions," he explains. "I can use the best possible truck to pick up a load without considering the driver, and then use the swap system to satisfy the driver's needs."

Today, with the help of optimization, the fleet has pushed on-time deliveries to 98% and lowered emptymiles by 2% in a largely regional operation.

Despite the positive results, Johnson believes that fleets need to keep their eyes open when considering optimization. It took M.S. almost a full year to build its preference tables, which could have been problematic since trucking is a dynamic business that is constantly changing.

"Most fleets have changed drastically over the last four years," he says. "We're doing a lot more regional business these days. If you're not changing your tables to reflect that, you're going to start getting recommendations that don't make sense."

Theoretically, optimization should allow M.S. to handle more drivers without increasing the number of dispatchers. "In reality, that hasn't happened," says Johnson. "I'm sure it will, but not for a while. It can take two or three years before you start to see a real return on investment with these systems because you have to couple the tools with changes in things like corporate marketing efforts. It's not just technology. It also involves people and processes. You won't see the return until you're able to truly integrate these systems into the fleet's workflow."

Now that it's at the top of a three-year learning curve, M.S. "is really, truly optimizing freight," says Johnson. "We're making good decisions about every piece of the delivery chain because these tools have become part of our culture and our processes."

Johnson's advice for truckload carriers considering optimization tools is to start with an honest evaluation of your current operation. Why do you need it? Are you good at what you do? How much better can you get? How many more miles can you actually run? And most importantly: Are you willing to make the required effort to execute true optimization, or are you kidding yourself about your resources and commitment?

Resistance in the ranks Prime Inc. installed both strategic and operational optimization tools a little over three years ago, but up to this point they've seen limited use, especially the load-matching system. The operational system recommends loads that Prime's leased drivers can either accept or decline. It also suggests trailer swaps, which again can either be accepted or declined by its owner-operators.

Use of the load-matching system has gone up and down, with the swapping option used most heavily around the holidays to get drivers home, according to Karen Brigham, director of technology. The biggest hurdle hasn't been implementing the system, but acceptance by load planners, especially the fleet's experienced planners, she says. Load planning is a fast-paced, complex process, not unlike a game of chess or bridge. Asking an experienced person to hand over that portion of their job to a computer can generate a good deal of resistance and faultfinding.

"If you look at a single load, you may be able to question the (system's) recommendation, but the issue is planning 200 loads," says Brigham. "There's so much data that has to be added to the mix that no one person can make all those decisions."

Prime doesn't require its planners to use the optimization systems, but they have been expected to handle an increasing number of loads.

"Now we're getting more planners, especially the younger ones, asking for access to the system," says Brigham. "They're finding out how useful it can be, especially if a driver turns down a load and they have to reoptimize an area quickly.

"Optimization is one of the ways we're going to leverage our investment in sophisticated systems so that we remain a leader in our markets past 2000," she says. "Our goal long-term is to have operations people deal with drivers and customers and leave the load planning to the machines. I don't know how soon that will happen, but it will."

Prep work pays Sitton Motor Lines has only been using its load optimization system for 18 months, but it's already seeing dramatic improvements in equipment utilization, customer service, and driver satisfaction. Deadhead miles for the 650-tractor truckload carriers are down 21%, while on-time deliveries have hit 99.2%, according to Ron Lazo, senior vp. Better still, driver turnover has dropped from an annual rate of 112% to 59%, primarily because optimized trailer swapping has allowed Sitton to meet driver requests for time home, he says.

The key to such rapid results was proper preparation for optimization, according to Lazo. In fact, it sounds as if Sitton had worked through Larry Johnson's list of questions.

"Three months before we began implementing the system, we raised the idea of optimization (with employees) and started training them even before the system was installed," Lazo says.

Since operational load matching systems are working in real time, the data feeding those systems has to be good. "We knew that data integrity was the key to making optimization work, and the truth was that we hadn't been doing the best job in that area," says Lazo. "So three months before we installed the system, we also went to work on cleaning up our data (acquisition) so that we'd be ready."

With both the personnel and processes prepped, the actual implementation took less than two months, and Sitton began seeing results almost immediately with one exception. Initially, drivers weren't happy with "a computer running my truck," says Lazo, but the benefits of the system, especially its ability to keep them loaded and to get them home when they wanted, quickly won them over.

While its initial efforts with optimization focused on operational systems, Sitton is currently working with a vendor to develop a more powerful strategic tool. Moving beyond historical analysis of profitability, the new system will add yield management tools not unlike those used by airlines to manage fair prices for maximum profit.

"It's actually both a strategic and a tactical management tool," says Lazo. "It should be very valuable since trucking operates in such a dynamic environment."

More for less Load optimization becomes far more complicated when it moves into LTL operations. Instead of relatively straightforward power-to-load matching, the systems have to handle fluctuating freight volumes moving through a fixed set of terminals connected by an almost infinite number of possible traffic lanes. Add pickup and delivery to the mix, and it becomes easy to understand why LTL optimization was originally limited to only the largest national carriers with immense information technology resources.

That's changing. Computer technology has gotten less expensive and easier to use, bringing new capabilities to smaller, regional fleets just as they've begun to see themselves as successful competitors against the large national carriers. Use of optimization still isn't universal among LTL carriers, but many of the most aggressive regionals are now using both strategic and operation systems to help them maintain that competitive edge.

Southeastern Freight Lines has a fleet of 1,700 power units and 5,300 trailers spread among a network of 49 terminals that stretches from Virginia south to Florida and west to Texas. "I guess we could run the fleet without (optimization) tools, but I don't know if we could do it for long," says Braxton Vick, senior vp-planning and development.

Three years ago, Southeastern began using a system to help it make decisions on inbound trailer placement at dock doors. Since that time, "we've dramatically increased our use of decision support tools," says Vick. "Today we can analyze an account for profitability, identify what's making that account unprofitable, and even produce reports for our customers telling them what parts of their operations are causing problems."

On the operations side, Southeastern uses a routing system that plans each shipment's movement from origin to destination, including the local delivery portion. The fleet is also in the process of developing a central dispatch system that will help it optimize linehaul operations on a real-time basis rather than a day at a time.

In addition, the fleet is testing a wireless communications system on pickup routes in three locations. The system, which equips drivers with handheld computers, will let the fleet's other optimization tools begin building outbound loads hours earlier and with better information integrity.

Strategically, the fleet optimization systems project volume models for the next five years so Southeastern can plan to meet equipment, terminal, and yard capacity requirements. "We rerun that plan at least twice a year," Vick says.

Looking at a much shorter timeframe, the fleet also produces reports that graphically depict the previous day's activities for executive management. "On a daily basis, we're able to see trends based on actual data," says Vick. "Most fleets don't have that kind of information until the end of the month.

"We stepped up our use (of optimization tools) three years ago because we needed to get better," says Vick. "Our customers wanted us to do more, and our competition was also getting tougher. You can only get so far reacting to what someone thinks is best. In the end, someone still has to make the decisions, but it always helps when you can bring the facts to the table, when you can see what's really going on under all that data."

In recent years, many fleets, both for-hire and private, have invested heavily in advanced information systems and wireless communications. As they get better at gathering data, however, the value of that information can easily be lost behind the sheer volume. As some fleets are finding out, optimization systems can actually take advantage of those mounds of data to suggest actions with real and positive effects on fleet operations and profits.

Hide comments

Comments

  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Publish