Mathematics of Planet Earth

As more and more purchasing takes place online, I’ve been wondering whether it’s more energy efficient to go out and buy something at a local store or to order it over the internet and have it delivered to my door. And which one has the smaller carbon footprint? Now it’s pretty simple to figure out my cost in time and money, and so like millions of other people I often decide that for me online is cheaper. But I see the cardboard boxes and the packing material filling up the recycle bins where we live, and I notice the delivery trucks every day making deliveries on our block, and I wonder about the differences in the total energy costs of the systems for getting goods from manufacturers to customers.

Well, I have found some studies of exactly these questions, and the answer is: it all depends. But what it depends on is something easy to analyze and to a great extent something that I can control. That key factor is the trip from home to store—how far it is and how I get there.

This conclusion comes from a study of the logistics of delivering flash drives from the manufacturer to the customers’ homes by traditional retail and by online retail. Not included in the study were the energy used and CO₂ emitted for the pieces of the supply chain that both had in common such as the manufacturing process, and so it is not a total accounting of energy consumed and carbon footprint. And although the study focuses on a narrow product line, the data used came directly from an online seller and a wholesale supplier. (The research appears in Life Cycle Comparison of Traditonal Retail and E-commerce Logistics for Electronic Products: A Case Study of buy.com. The authors are C. Weber, C. Hendrickson, P. Jaramillo, S. Matthews, A. Nagengast, and R. Nealer from the Green Design Institute, Carnegie Mellon University.)

Not surprisingly, the packaging costs are significant for e-commerce and hardly important for traditional retail, and the computer network costs are much higher for e-commerce, but what surprised me is just how big a factor the “customer transport” parameter is for traditional retail, accounting for about 65% of the energy consumed and the CO₂ emitted. The corresponding parameter for e-commerce is the “last mile delivery,” and although it is even bigger than packaging, it only comes to about a third as much, on average, as customer transport. Here is a chart from the study showing the makeup of the carbon footprints.

CO₂ emissions associated with retail and e-commerce delivery systems by stage

Using probability models derived from the data for the parameter values, the authors ran Monte Carlo simulations to draw conclusions. They estimate that only 20% of the time does going to the store to make your purchase result in less CO₂ emitted than having it delivered.

Now what can I do about all this? Well, there is great variability in the two most important parameters: distance to the retail store and and the fuel economy of the customer’s car. There is so much variability, in fact, that if I walk to the store, it is almost surely uses less energy and emits less carbon dioxide than ordering online. If I drive just a couple of miles in a car that gets average mileage it is still likely to be less energy intensive. And if it is just a couple of miles, then I might ride my bike instead.

Kent E. Morrison
American Institute of Mathematics