Vertical farms today waste space and energy because their fans and ducts force wide gaps between plant racks and cannot control the microclimate around each plant precisely enough to optimise growth. This £385,689, 18-month project will replace those fans with Acer’s Environmental Control System, which separates the air around each plant canopy and manages a much smaller volume of air, giving precise control of temperature, humidity, and CO₂ at every location in the farm. The problem is that current vertical farms cannot compete with mass-market supply chains on productivity. By doubling the number of rack layers and increasing yield per metre by more than 20 percent, Acer’s technology aims to make vertical farming commercially viable for packed salads and herbs at scale. A machine-learning simulation, built with AI engineers at PhysicsX and validated at a lab-scale demonstrator farm run with Professor Tracy Lawson at the University of Essex, will generate the training data needed to design low-energy cultivation recipes. If successful, a single farm could produce 3–4 tonnes of fresh produce per day, with eight farms generating £80–£100 million in annual revenue by 2035. This would cut food miles, water use, and pesticide reliance without sacrificing cost or output.
View original technical description
Vertical farming brings **significant benefits** to fresh produce production including reductions in waste, carbon emissions, water and pesticide usage, and food miles. These are possible because vertical farming can **precisely control** all aspects of the cultivation environment (light, temperature, humidity, CO2, etc.). However, the current generation of farms cannot deliver the productivity to compete with existing mass-market supply-chains at the scale required. This £385,689, 18-month industrial research project will overcome the key limitation to current vertical farms' ability to compete effectively; a reliance on fans and ducts to control temperature, humidity, and CO2, which results in **large spacing between racks** and an **inability to deliver the precision required to optimise the cultivation process.** This project will use a state-of-the-art, high-fidelity machine learning simulation to develop and demonstrate outstanding cutting-edge technologies (Acer's Environmental Control System) that separate the air around the plant canopy, reducing the volume of air managed, and enabling precise control of the microclimate around all plants, in all locations within the farm. This project is only made possible by leveraging **Acer's unique experience** in large-scale, commercial vertical farming in Asia, and by bringing together world-class experts. Acer will collaborate with Professor Tracy Lawson at the University of Essex to develop low-energy cultivation processes in a lab-scale Demonstrator Farm, evidencing the productivity that Acer's technologies make possible at large-scale, and generating real-world training data for the simulation. Acer will work with the global-leading AI simulation engineers at consultancy PhysicsX to deliver the simulation. Precision control of the cultivation environment enables Acer to: 1. Minimising the space between plant and light, **doubling(2x)** the number of rack layers compared to the current generation of farms. 2. Deliver novel, low-energy cultivation recipes that maximise yields and minimise costs, **increasing productivity per metre by a further \>20%**(Kumazaki, 2022). Acer is committed to **innovation** and the development of commercial activities **in the Cluster**, and is planning to deploy the technologies developed through the project to grow packed salads and herbs, first at small-scale, in 2026\. Once productivity has been demonstrated on a commercial basis, a mass-market scale farm growing **3-4 tonnes/day** can be built. The size of existing demand means that once production has been stabilised, 1-2 farms/year can be added. Three farms with Acer's technologies are planned by 2030, with 8 farms generating annual total revenue of £80-£100m by 2035\.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know