Lean manufacturing for indoor agriculture
Managing throughput and capacity with aquaponics vs hydroponics
--
Within the world of indoor agriculture, aquaponics is a niche player compared to its cousin, hydroponics. If you’re a traditional greenhouse operator or just a hobbyist, there are reasons to choose hydroponics over aquaponics. Information on aquaponics is buried in academic literature while hydroponic handbooks abound. Aquaponic operators need to devote valuable farming real estate to fish and fertilizer production while hydroponic operators don’t. More directly, fish feed is more expensive than hydroponic fertilizer.
As one of the few tech-focused, commercially-driven aquaponic companies in the world, we have a strong opinion: the future of large scale indoor farming lies with aquaponics, not hydroponics. The reason has to do with the increasingly manufacturing-oriented nature of indoor farming.
Whereas traditional farms’ primary concern is weather, indoor farms can control the climate. Instead of weather, indoor farms need to worry about capacity and throughput — the two problems that vex every manufacturer. In this post, we’ll explain how aquaponics is better equipped than hydroponics to manage capacity and throughput, drawing parallels to Toyota vs GM.
The capacity vs. quality tradeoff in hydroponics. Hydroponic produce is infamous for tasting like water. It looks great, it feels great, but the taste just ain’t there. High-end restaurants — a bellwether of food trends — have historically avoided hydroponic produce as a result.
The reason traditional hydroponic produce tastes bland is that traditional hydroponic growers fertilize all of their crops using the same chemical nutrient solution, despite the fact that different crops (e.g. arugula, basil, or kale) require different fertilizer formulations (e.g. pH and nutrient ratios). Given the number of variables to be controlled, most hydroponic growers choose to reduce complexity by resorting to the lowest common denominator — a single nutrient solution that is sufficient for all their crops but isn’t optimal for any of them.
That’s changing with the new generation of hydroponic growers like Gotham Greens and Aerofarms, who are able to achieve better quality and higher yield by using unique nutrient “recipes” tailored for each crop. With higher quality, this new breed of hydroponic grower is seeing wider adoption of their product, from chefs to high-end grocers, like Whole Foods Market.
There’s a tradeoff though — in order to deliver each crop its custom hydroponic nutrient recipe, separate fertigation systems (fertilization + irrigation) are required, and each fertigation system has finite capacity. Here’s an oversimplified explanation of the issue for precision hydroponic growers.
Let’s say you’re setting up a facility to sell 200 cases per week on average across two SKUs, spinach and arugula. Your anchor customer wants 80–120 cases per SKU per week, depending on the week. So what do you do? You set up two fertigation systems, each with a peak capacity of 120 cases. You grow at full capacity (240 cases per week), and any product that your anchor customer (200 cases per week) doesn’t buy gets sold at a discount to a distributor, donated, or thrown away.
The challenge for indoor growers is having enough capacity to accommodate variability in demand, without (1) growing product they can’t sell and (2) compromising quality and yield.
Learning from lean manufacturing. This multiple-SKUs-with-finite-capacity problem is a familiar one for manufacturing companies. Toyota became the most valuable automaker in the world by solving this problem most efficiently. Put simply, while US automakers, like GM, built a different production line for each car model or shut down existing production lines to retool for different models, Toyota created a single production line that could make any mix of their models on a given day.
The US automaker approach to manufacturing different car models was motivated by the idea that building multiple cars in a row of the same model was more efficient than building different models back to back. In terms of raw throughput, they were right. But this created an inflexible production model — given the time it took to create a new production line or retool an existing production line to accommodate demand for different models, the GMs of the world were less capable of producing to match variable consumer demand. The result was an excess of both parts inventory that had no business going into the production line and finished goods inventory that couldn’t be sold.
Meanwhile, Toyota created a single production line that could build different models back to back. Since customer orders for various car models change every day, Toyota was better able to match their production schedule to customer demand. If throughput was measured based on cars produced and sold rather than just cars produced, Toyota’s model beat GM’s by a longshot. As a result, Toyota’s model became the basis for the “lean manufacturing” philosophy and was subsequently adapted to other industries by companies as far ranging as Nike and Intel.
The key to Toyota’s process is producing “only what is needed, when it is needed, and in the amount needed” through two key innovations: just-in-time (JIT) manufacturing and the Kanban system. You can think of JIT manufacturing as the what and Kanban as the how. The point of JIT manufacturing is that if you’ve got customer orders for five Camrys, you have inventory for five Camrys. The Kanban cards (literally “signboard cards”), as illustrated below, are used to communicate when parts on the production line are depleted, replacement parts are retrieved, and travel back through the supply chain to control the production of new parts as needed. As a result, there is minimal idle inventory and finished goods — everything is tied to a customer order as the production line switches from one model to the next with frequent but short interruptions to the production flow as the production line accommodates different models.
Indoor growers are, for the first time ever, able to apply this methodology to the salad market and reap the benefits due to two recent changes:
- Drastically reduced cycle times. Lean manufacturing tactics are useless if you’re a traditional soil farmer with a 30 day cycle time — too long to make planting decisions based on incoming customer orders. Thanks to LED lighting, however, indoor growers can now achieve cycle times of 7–18 days — short enough to sync production schedules to accurate order forecasts. In doing so, indoor growers improve top line revenue by reducing supply disruptions and increase margins by reducing waste.
- Increasing fragmentation in the types of greens people eat. Twenty years ago, Americans ate spinach, iceberg, and romaine lettuce. Growers specialized in one or two of them. Lean manufacturing principles gain strength as product lines grow, and that’s exactly what’s happening in the marketplace. Supermarkets not only are stocking 8+ types of greens, but providing different mixes of these greens. If Whole Foods asks for 200 cases of a peppery mix of radish and arugula greens, delivering your arugula and spinach mix ain’t gonna cut it.
Why aquaponics is lean manufacturing for indoor agriculture. Aquaponics alleviates hydroponics’ capacity problem by using a single fertigation system, with a single nutrient recipe, while maintaining best-in-class yields at top-chef quality. The key is a bacterial kanban system.
Similar to how Toyota’s kanban cards regulate how resources are introduced on a just-in-time basis, plant growth promoting bacteria (PGPB) can decide what nutrients the plants need, when they need those nutrients, and how much is needed, moment by moment. PGPB attach themselves to or even grow inside plant roots and detect different metabolites (signaling molecules) emitted by plants depending on their physiological needs. The PGPB block or facilitate the uptake of different nutrients and even manufacture other compounds that plants need, like vitamins or growth factors.
This is huge. Without PGPB*, human operators are forced to guess what nutrients plants need, how much they need, and when they need it. Even plants of the same crop type have slightly different nutrient uptake patterns — they’re living beings with different genetic makeups. If each plant were a person, imagine trying to feed the same diet to thousands or millions of individuals and having them all thrive.
With PGPB, not only do you get optimized nutrition for each crop type, but you get it for each individual plant without any human intervention. A grower can get great yields and delicious product from multiple crops growing simultaneously — no need for different production lines, or in this case fertigation systems, for each crop.
PGPB have lived in symbiosis and co-evolved with plants over thousands (or millions) of years. Their ability to survive rests on their ability to detect and regulate plant nutrition with resolution and on a time scale that’s not possible for human operators. Studies pitting plants grown in an aquaponics solution versus an industry standard hydroponic solution bear this difference out.
Academic data aside, if you don’t believe bacteria does a better job at promoting plant growth than direct human intervention, take it up with Monsanto. Monsanto launched a joint venture with Novozyme, a microbial R&D firm, to study and develop bacteria-based plant treatments, and is betting its sales forecasts on them. It expects its latest corn microbial, currently in 10–20% of US corn fields, to be in over 90% of US corn fields by 2025 due to its ability to boost yield. That’s quite the turnaround for a company that built its $50B market cap on human-directed chemical intervention.
What this means for the business of indoor growing. The tradeoff for hydroponic growers is a challenging one: a single hydroponic nutrient system compromises quality and yield, but individualized growing systems for each crop requires building excess production capacity to maintain required output. Accordingly, as farming moves from the field to the greenhouse to the warehouse, lean manufacturing principles will need to be designed into farms from the ground up. We think aquaponics and the microbiome it supports is the right tool for the job.
What we’ve left out of this post is the quantified business case in choosing to operate an aquaponic system instead of a precision hydroponic system. In our next blog post, we’ll discuss the tradeoffs and how the numbers play out in a business case.
* Hydroponics is not completely without PBPG. A limited amount of PGPB can be found in hydroponics due to the fact that these bacteria thrive on organic matter, which is abundant in aquaponics but only available in trace amounts in hydroponics.
