Reflections on Oxygen Reduction Systems
This week I had the pleasure of attending a small conference on fire safety in warehouses. The conference was hosted by Wagner Group GmbH of Hannover, Germany (not to be confused with the Russian organization of the same name, which also deals with fire, but with different technology and intent). Wagner is a supplier of fire prevention systems based on oxygen reduction technology. They provide solutions for (i.a.) data centers, libraries and warehouses. Their primary approach is to prevent fire, rather than extinguish it, by reducing the oxygen content in the atmosphere to a level that does not allow most materials to catch fire. Often, this level is just below 15% of oxygen content in the air.
Conference attendees included representatives from authorities, insurance companies, real estate brokers, architects, civil engineering consulting firms and logistics consulting firms. The mix of participants made for a fruitful discussion and a variety of aspects were addressed.
Before you read on, I want to emphasize that this article was not sponsored or otherwise endorsed or written to please anyone. Instead, it is my personal summary of the above-mentioned conference and previous meetings I have had with fire protection company representatives. Any comments, corrections or additions to this article are welcome.
Types of Fire Protection Systems
When it comes to fire protection systems in warehouses, there are, roughly speaking, two types of systems. Sprinkler systems as the predominant solution and – less popular – oxygen reduction systems. (There is a third, carbon dioxide sprinkler systems, but the short version is: they belong in data centers, not warehouses. Don’t). While the role of sprinkler systems is to contain the fire until the fire department arrives and extinguishes it (note: it is not the role of the sprinkler system to extinguish the fire itself), the role of oxygen reduction systems is to prevent the outbreak and spread of the fire. There are cases where the outbreak of fire cannot be prevented (e.g., when lithium-ion batteries self-ignite), but the fire would not spread to pallets, plastic containers, cartons, other stored goods, etc.
Why is Oxygen Reduction Still a Niche Solution?
One of the questions discussed during the conference was why oxygen reduction has received comparatively little attention in the marketplace. This was the starting point for an exciting discussion. It seemed to me that the consensus of the participants was that confidence in the concept and technology of oxygen reduction was not one of the reasons for the low market penetration.
In my point of view, there are eight predominant causes of low market penetration at this point, and I believe they capture most arguments brought up during the conference.
- Uncertainty in total cost of ownership
- Insufficient understanding of the benefits
- The incentive structure in the buying center
- Lack of available data for quick cost estimates in the planning phase
- Standard buildings come with sprinkler systems
- Sprinkler systems are needed anyway
- Contracting effort for the customer
- The naturally slow diffusion of new technologies
Uncertainty in Total Cost of Ownership
Stakeholder: Customer
With any major investment, you should be aware of the cost profile and business case of the various options you have.
For a sprinkler system, the initial investment can be significant, especially since additional costs may be necessary, such as a retention basin for extinguishing water. But once it’s there, the ongoing costs are negligible.
With an oxygen reduction system, the profile is quite different. The initial investment can be much lower than a sprinkler system (not always, but sometimes). However, the ongoing cost is much higher because energy is required to maintain high nitrogen and low oxygen levels in the atmosphere. And of course, where energy is involved, there is the uncertainty of energy prices: What if government idiocy has not peaked yet and the price of electricity will be even higher in the future? Also, for an oxygen reduction system to work, the building must be well sealed. What is the long-term experience with this sealing? Will it need maintenance at some point? How does the economic life of the sealing compare to the economic life of the warehouse? These questions are difficult to answer, so there is a lot of uncertainty in the total cost of ownership.
Insufficient Understanding of the Benefits
Stakeholder: Customer
Not only are life cycle costs of oxygen reduction systems poorly understood, but so are their benefits. With other words, the complete risk profile is not understood.
The purpose of sprinkler systems is to contain fire until the fire department arrives. Neither do they prevent it, nor do they extinguish it. Depending on the system setup, the sprinkler system may work effectively or poorly. Imagine a fire breaks out on a low level of a cube storage system. Several tests (documented on video, some publicly available and others confidential) have shown that the time for the sprinkler system to respond and its effectiveness once it does respond are inadequate. And when a large fire is sprinkled, it creates large amounts of smoke that contaminate other products, which can make them unsaleable, even if they were not damaged by the fire itself.
While the insurance would cover the material damage and perhaps even the lost profits due to the business interruption, it cannot prevent the customers from not being able to wait for the warehouse operator to rebuild the system and start operations, so they will have to look for new suppliers – almost certainly some of them will never return to their old supplier. This is an impossible risk to insure against. Anecdotally, it takes an average of three years for a company to file for bankruptcy after its warehouse burns down. That makes sense when you consider how long it takes to clear the site, rebuild the facility, resume operations and find that customers are gone. What price do you put on the survival of the business? How do you factor this into your business case calculation?
This is exactly the situation you could most likely avoid with an oxygen reduction system. “Most likely” – because according to FM Global, the available data is insufficient for statistical analysis. However, it is fair to say that the risk of warehouses with reduced oxygen levels in the atmosphere burning down appears to be much lower than of conventional warehouses with sprinkler systems. I am not aware of any case where this has actually happened, but I would appreciate an email if you have other information. Even if the total cost of ownership of an oxygen reduction system were higher (and sometimes it is, sometimes it is not), if in the event of a fire, the business interruption would be minimal and the business would continue, this seems to be a major benefit, and one that I would expect would not be included in most business case calculations.
Also, insurance companies do not seem to reward risk reduction. Neither do users of oxygen reduction systems benefit from a discount on their fire insurance policies (but neither do users of sprinkler systems), nor can customers be assured that it will be easy to find an insurance company in the first place. One conference participant pointed out that insurance companies have not been able to make money in the warehouse sector, so rising rates and greater reluctance to provide fire insurance are to be expected. One example discussed in this context was a warehouse that was equipped with oxygen reduction technology ten years ago and had no problems with it, but the insurer would not renew the policy. This lack of predictability can obviously be a big problem for the warehouse owner. As mentioned earlier, the same thing can now happen with a sprinkler system. In general, it seems to have become much more difficult to get fire insurance on certain systems (mostly cube storage systems) since two Ocado warehouses burned down in Andover (2019) and Erith (2021).
Incentive Structure in the Buying Center
Stakeholder: Customer (decision maker at the customer)
For the decision maker in the buying center for any technical solution, there is a clear incentive to play it safe. Not safe for the company, but safe for him personally. As an employee of a company in a trusted position, you obviously care about risk to the company and you want to make the best decision for the company. But even more important to you is that you can pay your own bills and feed your family. So the decision maker will most likely minimize the risk to himself. He will play it safe. And that means he will choose the technology option with the greatest penetration and acceptance, even if there is a better but less accepted alternative. “No one has ever been fired for buying an IBM”.
Lack of Data for Quick Cost Estimates During Planning
Stakeholder: Warehouse automation company
The bidding phase for warehouse automation projects often begins with a request from the customer for a budget proposal. Because planning time is expensive, available time is short, workloads are high, and the chances of winning the project are less than the chances of losing it, warehouse automation companies try to keep budget proposal efforts in check. Planning engineers like to work with estimated values so they don’t have to obtain firm quotes from suppliers. This often happens with cost items such as roof and wall, IT, project management – and fire protection. Because of the large number of completed projects involving sprinklers, planning engineers may use estimated cost values based on past projects of similar scope. They cannot do this for oxygen reduction systems due to lack of available data.
Contracting Effort for the Customer
Stakeholders: Fire prevention technology provider, customer
For obvious reasons, customers want to reduce (a) the risk and (b) effort in the procurement process of hugely expensive assets such as warehouse automation technology and fire protection systems. From the customer’s perspective, in many cases it would be ideal to work with a general contractor who takes care of all subcontractors. This would allow the customer to deal with only one party for contract negotiations, project management and scheduling, warranty claims, maintenance, etc. Large warehouse automation companies such as SSI Schäfer, Dematic, Witron or Swisslog often include all works in their scope of supply, including sprinkler systems. It seems less common for oxygen reduction systems to be included in the warehouse automation vendor’s scope of supply (with Jungheinrich being a notable exception with its PowerCube system). Worse, even oxygen reduction system vendors don’t seem to include necessary components such as sealing and airlocks in their scope of supply, resulting in more effort on the part of customers willing to try the technology.
Standard Buildings Come Sprinkled
Stakeholders: Real estate developers, 3PLs, customer
Many warehouses on the market adhere to de facto industry standards. They are no higher than 12 meters, have either a maximum area of 10,000 square meters or multiples thereof, and have sprinkler systems installed under the ceiling. And as long as such a building is only leased for a few years (the typical two- to three-year 3PL lease term), or if the owner of the building changes frequently, no one will install an oxygen reduction system. Similar to most warehouse automation solutions, oxygen reduction systems must be tailored to the specific needs of the customer. If the customer changes, it is unlikely that the system will meet the new customer’s needs. And why bother when the building is equipped with a sprinkler system anyway?
Sprinkler Systems are Needed Anyway
Stakeholder: Customer
Oxygen reduction is a fantastic technology for a cube that can be sealed, whether it’s a high-bay warehouse or an automated cube storage system. It is not an appropriate technology for a low bay picking area, a packing area, a staging area with truck gates, or the like. Neither would it be necessary or make financial sense. In those areas, you would install sprinkler systems. The larger these areas are, and the more sprinklers you need to install anyway, the lower the marginal cost of extending the sprinkler system to all areas of the warehouse. To be clear: We are talking about costs here, not benefits. The arguments made in the section on insufficient understanding of the benefits of oxygen reduction solutions well apply when the TCO of a sprinkler solution is lower than that of an oxygen reduction solution.
Naturally Slow Diffusion of New Technologies
Stakeholders: Insurance company, customer
A broader issue is simply the natural adoption of most new technologies, especially in high-risk areas, where diffusion is slow at first because most potential buyers are risk averse and await confirmation of technological and financial feasibility and prudence. Once a certain threshold of references in the market is crossed and the positive feedback loops kick in, adoption picks up. AutoStore is a good example of a system that had been around for 20 years or so without any notable market success, then gained momentum around 2015/2016 and is now the best-selling solution in the warehouse automation market (by a huge margin). While AutoStore’s example is quite spectacular and it is unlikely that oxygen reduction will become the predominant solution in most warehouses, the pattern of diffusion could be similar.
Conclusion
Fire protection for warehouses can be tricky, and it’s unlikely to get easier anytime soon. One of the changes on the demand side for fire protection systems is that the number of batteries in warehouses has increased dramatically and will continue to increase for many years to come. With the electrification of bicycles, society has managed to turn the most sustainable short-distance transportation system into one that requires batteries that are expensive and dirty to produce, have highly problematic working conditions for raw material extraction, have short life cycles, and require special handling for disposal. But that’s just in passing… What is important for the context of this article, however, is that lithium-ion batteries can suddenly self-ignite and burn even in reduced-oxygen atmospheres, making their storage in warehouses quite problematic and dangerous.
One of the lessons warehouse automation customers can learn from the discussion is that they should seek approval from regulators and insurance companies as early as possible – once a concept is in place and before a contract is signed with the automation provider. There seem to be many examples of warehouses that either have not received regulatory approval or have been unable to find an insurer.
Oxygen reduction seems to me to be a promising technology, and I would expect interest and adoption to increase significantly once the points raised in this article are addressed.
