Researchers from MIT's Center for Transportation and Logistics, with support from logistics services provider CH Robinson, have developed a new approach for calculating CO2 emissions for less-than-truckload (LTL shipments).
Of course a growing number of companies, both shippers and carriers, are finding the need to calculate C02 emissions from their supply chains, especially logistics operations.
The approach to doing so for full truckload shipments is fairly straight forward, most often calculated using something called the Greenhouse Gas Protocol Corporate Accounting and Reporting Standard (GHG Protocol), which was developed by the World Business Council for Sustainable Development and the World Resources Institute in September 2001.
It's a simple calculation, using one of two approaches. The first is to measure the amount of fuel that is consumed directly, multiplied by some emissions factor for the type of fuel used, generally diesel.
That approach works for carriers and shippers running their own fleets, but shippers using third-party carriers may not have access to fuel consumption directly. So, fuel consumption must be estimated using a proxy factor, such as vehicle miles traveled or ton miles moved.
There are some minor complexities even for truckload shipments, such as how to account for multiple stops on a given load, or whether to consider that trucks will get better mileage for lighter loads. For the latter, shippers likely just allocate the CO2 to individual shipments using weight or cube as the factor. We suspect the latter issue is generally just ignored.
But all of this is much more complicated for LTL shipments. Why? Two main factors: (1) There are substantial pick-up and delivery operations on the front and back end of the line haul moves between LTL terminals that the GHG protocol does not address at all; and (2) There can be as many as 30 or more different shipments from different companies with great differences in weight riding on the truck at all three stages (pick-up, line haul, and delivery), causing more complex allocation challenges.
With no good existing methodology, must companies trying to track CO2 from LTL shipments use the GHC protocol based on distance from origin to destination, which usually well underestimates actual emissions because the pick-up and delivery portions of the move are not considered.
The research was performed by Guilherme Veloso de Aguiar and Mark Anderson Woolard, graduate students in MIT's supply chain program, published as both a master's thesis and now in summarized form in a new white paper from CH Robinson. That company also helped the researchers get detailed network and shipment data from a major LTL carrier, in the end producing a data pool on some 2700 shipments that was used to develop the methodology.
For these shipments, CO2 emissions were calculated in two parts using the carrier's data:
Line haul emissions: The line haul distance was determined between the first and the last terminal that a shipment goes through. The total amount of fuel burned is calculated by dividing the total distance by the fuel economy factor (miles per gallon). The MPG is a parameter in the model based on one year of mileage and fuel consumption data, as supplied by the carrier. Using an emissions factor provided by the EPA Climate Leaders guidance material, the model calculated the total amount of line haul emissions.
The line haul emissions were allocated to each shipment, based on unit weight and the load factor, which was also derived from data provided by the carrier. Shipment cube, class, and load density information was not taken into account because weight data is generally accepted as more reliable.
P&D emissions: In the next stage of the model, emissions were added for the pickup and delivery segments of the shipments. Values for the number of miles covered and amount of fuel burned were assigned to each shipment based on respective origins and destinations. An MPG value was applied for the P&D operations, also based on carrier data, and the model computed the emissions from each activity.
The overall emissions number is naturally then equal to the sum of the line haul emissions and the P&D emissions. P&D, by the way, accounts for roughly 30% of the total CO2 emissions for an average LTL shipment, the research found.
That approach works great - if you have all the network and shipment details that a carrier does.
But a shipper will not have that detail, so how can it calculate CO2 for an LTL shipment? The researchers' answer was to develop a simplified model that requires just the same three inputs used for the detailed carrier network model: origin ZIP, destination ZIP, and shipment weight in pounds.
However, there are two main differences in the way the lower-precision model is structured.
1. Since the carrier's network is unknown, it is not possible to determine which terminals the shipments pass through. Therefore, the line haul distance is determined by a regression model based on the great circle distance between origin and destination ZIP codes.
2. P&D miles cannot be determined for each specific terminal. Instead, there are pre-determined values that change based on the origin and destination ZIP codes. All the other parameters (such as empty miles, MPG, emissions factors, and load factors) are equal to the values used in the detailed model. It should also be noted that these figures can be adjusted by the user.
All that leads to the formula described in the graphic below:
Source: MIT/CH Robinson
Of course a shipper using this formula can adjust parameters, such as MPG or average P&D miles, per their own needs.
The good and bad news: At an aggregate level, meaning measuring CO2 emissions from all or a large block of LTL shipments, the simplified modal is pretty accurate, returning emissions estimates that were just 3% off (higher) from what the detailed network model would have produced.
Unfortunately, in some specific cases, the simplified model yielded results that were as much as 3.5 times higher in emissions than the detailed model.
"This happened primarily because the simplified model utilized network averages from across all terminals, while the detailed model calculated emissions based on the characteristics of each terminal," the report explained.
So the bottom line: the relatively simple and flexible model for LTL shipments should work fine for most shippers across their LTL shipments in total. If great precision for emissions from individual shipments is needed, most likely that can only come for now from carriers themselves using something like the detailed model developed in this research.
More detailed information can be found here: Estimating Carbon Emissions from Less-than-Truckload (LTL) Shipments
What do you think of this LTL research? Was this an important CO2 measurement challenge? Let us know your thoughts at the Feedback button below.
|
