In this age of climate change and environmental awareness, individuals and businesses alike have been starting and continuing to evaluate their own role in the system. The wine industry is no different, and is often one of the first industries to adopt more environmentally friendly practices. Many wineries have started using environmentally sound viticultural and winemaking practices, and have made it clear to their customers that they are choosing to “go green” to do their part in protecting the environment. Recycling wine industry wastes have become increasingly popular as well, a topic of which is not lost on The Academic Wino. From meat preservation and nutrition enhancers, to animal feed additives and human disease amelioration, wine industry residues are frequently being studied for recycling purposes.
One frequently overlooked part of the wine industry, which is equally; if not more important than energy savings in viticulture or winemaking practices is the distribution side of the industry. In order for you and I to enjoy a wineries wines, there is inevitably energy costs involved, from the packaging to the storing, and finally to the transportation of the wine from the winery to your home. The study discussed today, though published in 2009, is still very relevant and provides very fascinating insights and analyses on the carbon and energy profiles (“footprints”, if you will) of wine distribution in the United States. Using a modeling system entitled CargoScope, an online tool that allows the user to build a supply chain network and define storage, transit, and processing parameters for each level, the authors of this article demonstrate that different types of distribution systems varying widely in their carbon footprints, while also making recommendations to wineries for lowering their emissions where it relates to the distribution of their wines.
Very little research has been done I regards to the outbound distribution of wine to consumers. The wine itself only accounts for 40% of the overall volume of a case of wine (standard 750mL bottles), and often needs to be stored in climate-controlled areas to avoid spoilage. This extra weight and cooling, in addition to the actual transportation of the wines, require attention in regards to attempting to lower a wineries’ carbon footprint.
How are wines distributed in the United States?
After prohibition, a three-tier system was put in place in the US, in an attempt to prevent overconsumption by requiring those that produced the alcohol to sell to retailers via distributors, of which all tiers are owned by separate companies or individuals. In some states, wineries are allowed to self-distribute, thus eliminating the distributor step. Many wineries are able to ship wines directly to their consumers, and of course, consumers can always purchase and pick up wine directly from the winery itself. Another spin-off of shipping directly to customers is through wine clubs, where customers sign up to receive multiple shipments of wine per year to either be picked up in the tasting room, or shipped directly to their homes.
|Figure 1 from Cholette and Venkat, 2009.
Illustrates how wine is distributed throughout the United States
Creating model distribution systems for analysis
In this study, the authors have chosen to examine a Sonoma winery that is trying to send wines to customers locally to San Francisco, and at a long distance to Manhattan. The criterion and information entered into the model were done so by consulting with individuals involved with each step of the process. The only step the authors were not able to find someone to consult with was the distributor/wholesaler step. If not data were obtained, assumptions were instead made and applied to the model.
The model winery in this example was based off of Cline Cellars, which is a medium-sized California winery with moderately priced wines priced around $10/bottle which sell in stores nationwide, as well as higher-end wines priced around $25/bottle which are only available through direct-to-consumer avenues. The wines in the model were assumed to be available in standard 750mL glass bottles, with an average/typical order size per customer equally 6 bottles.
The chosen distribution warehouse for the model was Southern/Glazer’s, located in Union City, which in 2009 distributed over 80% of the wine and spirits sold in the US. A regional customer warehouse in Richmond was selected as an optional option, since the retail chain Cost Plus World Market has a large wine sale facility there. For direct shipping to the local market, the major third party logistics provider for California wines, New Vine Logistics, was chosen. For the long-distance market, Federal Express was chosen, since they have a large sorting and distribution center in San Francisco. For local retail stores in San Francisco, the model assumed customers lived approximately 3.6km away (which is below the national average of 10km). For the long-distance market, rail and air hubs were also included in the model, in order to compare these modes of transportation to long distance truck transport. As a side note: for the rail hub option, trains were required to travel out of the way through Chicago, since there is no direct line from San Francisco to New York that avoids detouring through this northern hub. All distances were calculated using GoogleTM Maps software.
Simply put, the two general models examined were 1) a Sonoma winery delivering 6 bottles of wine to a consumer in San Francisco, and 2) a Sonoma winery delivering 6 bottles of wine to a consumer in Manhattan. Listed below are the various scenarios that were created for each general model:
1) 3-tier distribution:
· Mid-sized trucks used to transport wine from the winery warehouse to the distribution warehouse. Assumption: 100% efficiency and utilization.
· No temperature-controlled storage needed, since time in warehouse is short.
· Assumption: wines spend a week at the distribution warehouse until a mid-sized trucks move them to the retail shop.
· Assumption: wines spend a week under temperature control at the retail shop before the consumer purchases it by driving in a Honda Accord (fuel efficiency of 9.8 liters per 100 km; utilization of the car = 24%).
o Scenario variants: A) consumer gets to the store without a car; and B) consumer arrives at the store with a car, but utilizes it more efficiently by loading it up with many purchases.
· Assumption: warehouses are highly utilized.
· How the consumer stores the wine once at home is not considered in this model.
2) 3-tier distribution via retailer warehouse
· The previous scenario was modified by adding an extra retailer warehouse step into the model before the wines ends up directly in the retail shop.
· Assumption: wine stays in this warehouse under temperature control for a week before moving into the retail shop.
· All other steps are the same as the previous scenario.
3) Winery self-distribution
· Since many wineries are too small to utilize a mid-sized truck (carrying capacity of 344 cases of wine), this scenario assume uses of a light truck (carrying capacity 33 cases of wine).
· Removing the distributor level decreases storage time/costs by one week, and slightly decreases the total distance traveled.
4) Fulfillment via third party logistics (3PL)
· In this scenario, wine is shipped to customers through direct sales channels.
· Mid-sized trucks from the 3PL pick up the wine from the winery warehouse, and bring it to the 3PL temperature-controlled warehouse.
· A small package carrier picks up the wine from here in a mid-sized truck and brings it to their sorting center (not temperature controlled) every two weeks.
· A light truck brings the package directly to the consumers house from this sorting center.
5) Consumer picks up directly from the winery
· Only the fuel used for the round-trip is considered for this scenario.
· Assumption: the same Honda Accord from Local scenarios 1 and 2 is used.
oTwo scenario variants: 1) a hybrid car is used and 2) the consumer picks up wine for not only his/herself, but for other friends/wine club members as well (isn’t that nice!). This scenario assumes the consumer is driving a mid-sized pickup truck that is smaller than a light commercial truck, and holds about 33 cases of wine.
1) 3-tier distribution
· This is very similar to the local 3-tier distribution scenario, however, now there is an added heavy-duty diesel truck (temperature controlled) to make the cross-country trek from the distribution warehouse in San Francisco to the sister distribution warehouse in New Jersey. Assumption: 100% capacity and utilization.
· Assumption: wine moves from the distribution warehouse to the retail store in Manhattan via a mid-sized truck.
· Assumption: consumers live about 0.8km from the retail shop and travel via foot or public transportation.
· Assumption: if walking to the store, the model includes the use of a cab on the return trip (since 6 bottles can be awkward and heavy to carry for some).
2) Fulfillment via 3PL ground delivery
· This scenario is very similar to the local 3PL scenario, except when the wine is about ready to leave the 3PL warehouse, it is packaged in multi-day, and temperature-regulated packaging for the long trip across country (requires greater energy).
· From the FedEx facility in San Francisco, the wine is transported via a heavy duty diesel truck to the FedEx facility in New Jersey.
· From NJ, the wine it brought directly to the consumer via a parcel truck.
3) Fulfillment via 3PL airfreight
· This scenario replaces the heavy duty diesel truck with a FedEx air cargo plane.
· Since the trip is relatively short, temperature-controlled packaging is not used.
· The FedEx facility is assumed to send a Manhattan-bound parcel truck to the airport to pick up the freight, since the two facilities are very close.
4) Fulfillment via 3PL rail
· This scenario replaces the heavy duty diesel truck/cargo plane with the railways (CSX).
· A mid-sized truck picks up the wine from the FedEx facility in San Francisco and brings it to the railway. Once at the FedEx facility in New Jersey, it is assumed to not linger long and be sent to the consumer via a parcel truck.
· The packaging used is considered the same temperature-controlled variety as the 3PL ground delivery option, since the travel time is longer.
The authors of this study present detailed results from each and every one of these 9 scenarios (5 local, 4 long-distance). In the interest of time and space, I will only present the results that showed the comparison across all scenarios. If you’d like more details about any one scenario in particular, please don’t hesitate to ask and I’ll be happy to supply those for you.
|Table 13 from Cholette and Venkat, 2009.
Shows scenarios ranked from the most energy efficient scenario (lowest emissions)
to the least energy efficient scenario (highest emissions).
- When comparing all of the scenarios, significant differences in emissions are prevalent.
- The least efficient scenario was driving a car straight to the winery, which resulted in 80 times the emissions that would occur if the local delivery were handled via a 3PL scenario.
o The least efficient scenario involved the one with the least distance traveled, indicating that the utilization of the vehicle is almost always more important than the physical distance traveled.
- The long distance 3PL viarail scenario had emissions equivalent to standard local 3-tier distribution scenario.
o Total emissions for the 3PL rail scenario were about 60% of those associated with truck transport, and about 8% of those associated with air transport.
The authors of this study make several recommendations to wineries, based on the individual scenario results. First, wineries should focus more on transportation of their wines as opposed to the storage and packaging of them, since the latter two do not significantly contribute to the overall carbon footprint of wine distribution. Wineries should stockpile their wines when possible, that way they are greater able to utilize the vehicles by which their ways are transported.
For long distance transport, the cross-country transit step contributes the most to emissions. Wineries should try and utilize the most efficient transport type possible, which according to these results, is the railways or airlines. For the local transport scenarios, the step from the retailer to the consumer generates the greatest levels of emissions. If the consumer is able to walk or take public transportation, rather than drive to the store, this would be beneficial to reducing overall emissions. Since this may not be feasible for many consumers, wineries may wish to promote the use of 3PL’s instead, supporting a more direct-to-consumer approach, and a reduction in emissions.
Since wineries most likely do not wish to discourage consumers from visiting their tasting rooms, they may encourage consumers/wine club members to pick up wines not only for themselves, but for other members as well, in order to avoid the addition emissions from another underutilized vehicle. Volume discounts or discounts on shipping costs may also encourage consumers to greater utilize their vehicles, and reduce carbon emissions. Sometimes just by simply mentioning that the winery is taking strides to lower their carbon emissions and educated consumers on the inefficiencies of the transport link in the system, then the consumers themselves may be motivated to try and do their part in reducing the carbon footprint as well.
This study, of course, is not without its’ faults. It makes many assumptions that may or may not be applicable or true in real world situations. Any changes to these assumptions, and the results would be altered. For example, if the utilization rates were less than 100%, the overall emissions would get larger, however, the relative rankings of each scenario would likely not change, unless some vehicles utilization rates were uncharacteristically different from the rest for one reason or another.
It is also likely that these scenarios wouldn’t play over exactly the way they were set up in the model over and over again, and situations would be different depending on the particular day(s) the scenarios occurred. Any number of variables could make the trek different from day to day, thus potentially altering the results.
One final problem is that every winery is different in regards to how it distributes their wines. For example, some wineries out “in the middle of nowhere” will encounter much greater difficulty in using one scenario versus another, and perhaps the most efficient scenario in general is not the most efficient scenario for that individual winery. Models need to be run on individual wineries in order to determine the proper scenario that is most energy efficient and economically feasible for that specific winery.
In general, I thought this was an incredibly interesting study, and it really shed some light onto how energy efficient wine distribution is in the United States, and sheds some light onto how wineries in general may change their distribution methods to create a smaller carbon footprint. Of course, wineries should include this sort of analysis in their overall evaluation of their total emissions, but be aware that this distribution step is a huge contributor to their overall carbon footprint.
I’d love to hear what you all think! Please feel free to comment below! Again, if you need more details on specific scenario model results, please don’t hesitate to ask!
Source: Cholette, S., and Venkat, K. 2009. The energy and carbon intensity of wine distribution: A study of logistical options for delivering wine to consumers. Journal of Cleaner Production 17: 1401-1413.
I am not a health professional, nor do I pretend to be. Please consult your doctor before altering your alcohol consumption habits. Do not consume alcohol if you are under the age of 21. Do not drink and drive. Enjoy responsibly!