Data-Driven Horticulture & Facility Startup

value of data-driven facility startup vs. “industry standard”

A facility opening up without a full sensor-based data-driven startup/optimization will be leaving Millions of dollars of revenue on the table.

A facility opening up without a full sensor-based data-driven bringup and optimization will literally be leaving Millions of dollars on the table. — Example (15k Canopy SF, Tier 3 Indoor Grow)

Custom Sensors & Horticultural data processing + visualization platform

Off-the-shelf sensors are often not suitable for CEA applications, leading to inaccurate data and a false picture of the real plant climate. To capture high quality data for this detail-oriented industry, we had to develop our own in-house sensor, algorithmic processing and data visualization platform. We employ the right transducers, packaged specifically for CEA usage, processed and and visualized to better understand and optimize the horticultural environment.

identifying and fixing issues in grow environment

We use our proprietary tools to search for hard to see problems. We can identify and fix horticultural issues before they impact the yield of the facility, increasing the stability and overall output of the investment.

Thermostat not accurately sensing plant conditions, resulting in inaccurate CEA setpoints, suboptimal crop yield & quality

Setpoint mistakes can easily reduce facility yield, and at worst, even result in total crop loss.

Checking for adherence to SOPs to identify and fix deviations when they occur.

Environmental optimization/tuning

Conditions within a grow room can very considerably across an x-y-z spread, particularly for dense racked operations, wherein providing sufficient canopy airflow can be a challenge. By visualizing environmental condition variance across a room (or rooms) we can further optimize the setpoints and in-room airflow strategy, resulting in healthier, more productive plants for a higher overall product quality and yield.

A/B test data - Control vs. Experiment - Modified airflow strategy to improve plant environmental conditions

We can accurately measure exactly how much a plant is transpiring over time, down to the mL. This results in an unprecedented model of the plant’s current state for crop steering, as well as the resulting latent load on the HVAC system over time.

KPI (key performance indicators)

We establish performance baselines for the facility systems, to compare environmental conditions across time, rooms and even buildings. Once these vital metrics can be measured, captured and properly visualized, we can then optimize setpoints to result in increased consistency, plant health, quality and overall yield.

Heatmap illustrating setpoint accuracy - actual VPD vs. target throughout grow room, over time

Machine Learning and AI for Agriculture

ML algorithms need data. Lots of data. But often overlooked is that they need high quality data. With our in-house-designed CEA-specific sensors and proprietary ML and AI-based algorithms we can tune and optimize facilities far beyond the “industry standard” facility startup method.

Optimizing a dry room, visualizing the moisture loss over time, to improve the quality and consistency of the dried product. Dry rooms are too often overlooked, resulting in an inconsistent and lower quality product, even if the plant was grown in id — Optimizing a dry room, visualizing the moisture loss over time, to improve the quality and consistency of the dried product. **Dry rooms are too often overlooked**, resulting in an inconsistent and lower quality product, even if the plant was **grown** in ideal conditions.

We can detect even the slightest light leak, which could otherwise potentially compromise an entire photo-dependent crop.

We measure key elements nobody else does in CEA.