Transportation Modeling: Is There Ever an Average Day?
Most large shippers spend a great deal of time and money collecting, analyzing, and maintaining the data used to drive daily transportation planning and execution.
When codified and integrated into the shipper’s transportation system, this data becomes the organization’s transportation policy. This policy is comprised of lanes, modes, rates, service levels, capacity, and a multitude of other variables that must work in concert to drive daily decision-making.
The transportation policy defines the framework upon which all transportation decisions will be made. It needs flexibility to allow the transportation management system sufficient latitude to find savings, and needs to take into account the typical variability found within virtually every supply chain. This is where shippers can often find opportunities to improve existing processes.
Deterministic vs. Stochastic Strategic Planning
Many analysts still rely heavily on statistical averages, ignoring the inherent variability that underlies nearly every aspect of transportation. The data used in daily planning ultimately reflects the inefficiencies.
So why do most organizations develop their policy this way? The simple answer is that averaging is easy to comprehend and calculate.
Yet averages rarely reflect reality. For example, if half of a supplier’s shipments weigh 5,000 pounds, and half weigh 40,000 pounds, the average is 22,500 pounds—a weight the supplier never ships. A fundamental input in the transportation policy is wrong.
A deterministic planning tool can be effective in daily planning when values are known. However, strategic planning should utilize a stochastic approach based on calculated probabilities, not the forecasted certainties implied in the deterministic approach.
Stochastic principles are not new to supply chain planning. The concept of safety stock exists because we cannot predict customer demand, lead times, or order fill rates with any certainty. The latest transportation modeling technology allows for combined stochastic optimization and simulation.
How Does This Work in the Real World?
Let’s take the example of prepaid-to-collect conversion—which shipments should be prepaid and which should be collect?
Prepaid-to-collect is a long-term, strategic decision that is subject to significant levels of variability that shippers must consider.
These variable elements include:
- Order quantities (can less-than-truckload shipments be consolidated?)
- Freight rates
- Fuel prices (if no fuel surcharge is assessed, the shipper takes on all the fuel price volatility risk)
- Order frequency
- Transport lead times (is there time to consolidate shipments?)
A deterministic approach assumes order quantities vary little, and that the customer understands the rate and fuel surcharges used throughout the year.
Through stochastic optimization and simulation, variable data can be modeled in a way that makes the analysts more confident they are making the best decision—defined not by a snapshot of data, but over time.
Most organizations take a deterministic approach simply because it has always been done. However, long-term strategic planning must account for known variables to create an efficient transportation policy that reflects the reality of an ever-changing world.