I believe that we can make a small dent on this problem if we improve the technology that runs greenhouses. (My bias: I work for a startup that is working on this and I'm kind of obsessed with the industry.)
Here is what current greenhouse tech looks like: a human (called a grower) uses their intuition to adjust hundreds of settings that parameterise what the actuators should do in different circumstances. For example, 20 settings might parameterise a 'ventilation setpoint line', which determines what air temperature the vents should start opening at given a measurement of the solar radiation intensity from a weather station. All the different actuators (heating pipes, vents, screens, misting systems, irrigation systems) have their own set of heuristics that need to be adjusted by hand..then beneath that a low-level control layer (e.g., PI control) manipulating individual motors/valves.
It's crazy but the grower has to manually adjust the settings as the weather, energy prices and crop state changes. With better software/automation we can improve the control of the environment within the greenhouse and thereby increase yields [kg/m^2] and minimise energy costs.
If the farmers are producing more tomatoes per m^2 for the same input costs they can afford to sell them to consumers (via supermarkets) at lower prices and still make money. It won't solve all the problems but would be a step in the right direction.
Plants don't really give a shit. Implementing a convoluted tech solution in greenhouses for an additional 5% efficiency will never pay back for itself.
> (My bias: I work for a startup that is working on this and I'm kind of obsessed with the industry.)
I suggest you actually learn more about growing plants than optimizing 20 different atmospheric and HVAC parameters. Many of these recent startups folded because they couldn't figure out that growing lettuce indoors hydroponically doesn't actually make anyone any money.
> I suggest you actually learn more about growing plants
We have people with decades of experience operating 70+ Ha of greenhouses in the team, so I don't think a lack of knowledge about the industry or plants is an issue. Perhaps my comment gave you the wrong impression about us.
> they couldn't figure out that growing lettuce indoors hydroponically doesn't actually make anyone any money.
It is quite easy to refute that statement that nobody can make money growing lettuce indoors by looking at Gotham Greens, or Revol Greens (https://www.revolgreens.com/). What is true is there are a number of agtech startups who raised a lot of money, did not really know what they were doing and imploded, I can agree with you on that.
I'm not sure exactly what you mean by 'efficiency', so I am assuming you mean yield. Have you ever looked at the income statement for a commercial greenhouse? From the data we've seen from operators inside the industry a 5% increase in yield roughly double your profits (consider that the costs for energy, labour, supplies is fairly fixed, so increases in revenue then flow down to the bottom line).
Yes, if you mean how much profit does the farmer make I think you are spot on.
Pricing fluctuates with the season, variety but from the data I've seen a farmer in the US might sell to the distributor at $2/kg, but after paying all their costs (packaging, labour, energy, supplies, depreciation) the pre-tax profit might be around $0.16/kg - so 8%, and that's probably a pretty good year! It's a low margin, high volume kind of business.
Here is what current greenhouse tech looks like: a human (called a grower) uses their intuition to adjust hundreds of settings that parameterise what the actuators should do in different circumstances. For example, 20 settings might parameterise a 'ventilation setpoint line', which determines what air temperature the vents should start opening at given a measurement of the solar radiation intensity from a weather station. All the different actuators (heating pipes, vents, screens, misting systems, irrigation systems) have their own set of heuristics that need to be adjusted by hand..then beneath that a low-level control layer (e.g., PI control) manipulating individual motors/valves.
It's crazy but the grower has to manually adjust the settings as the weather, energy prices and crop state changes. With better software/automation we can improve the control of the environment within the greenhouse and thereby increase yields [kg/m^2] and minimise energy costs.
If the farmers are producing more tomatoes per m^2 for the same input costs they can afford to sell them to consumers (via supermarkets) at lower prices and still make money. It won't solve all the problems but would be a step in the right direction.