Most paddy fields in Malaysia are still managed the same way they were a generation ago: irrigate on a fixed schedule, apply fertiliser by habit, and check for pests by walking the bunds. It works, until it doesn’t. By the time a farmer spots a stressed patch of paddy or a moth population that’s gotten out of hand, the yield has usually already taken a hit.
Paddy is one of the most closely watched crops in the country for a reason — it sits at the centre of Malaysia’s food security planning. But the way most schemes are run day to day hasn’t caught up to that importance. Soil condition, weather, and pest pressure are three of the biggest levers on yield, and on most farms none of them are actually measured.
Farming by Eye Has a Ceiling
Irrigation is a good example. Paddy needs specific water levels at different growth stages — land preparation, tillering, flowering — and getting that wrong in either direction costs yield. Without a reading from the actual plot, the only real feedback a farmer gets is the appearance of the crop, which shows up weeks after the water management was already off.
Fertiliser is the same story. Soil salinity and nutrient levels — what’s measured as electrical conductivity, or EC — and soil pH both drift over time, and both determine how much of the fertiliser a farmer applies is actually usable by the plant. Apply a standard rate regardless of soil condition, and you either waste fertiliser the soil didn’t need or under-fertilise a plot that needed more.
Pests follow the same pattern, just with higher stakes. Brown planthopper and stem borer moth populations build up quietly before crossing a threshold where damage becomes visible. Once you can see the damage in the field, the population has usually already been building for days or weeks.
What Farming Blind Actually Costs
None of this is a knowledge problem — Malaysian paddy farmers and extension officers know exactly what moisture, EC, pH and pest activity mean for a crop. It’s a visibility problem. Nobody can physically check every plot, every day, for conditions that are largely invisible without instruments.
The result shows up as yield variability between plots that should perform the same, fertiliser spent on soil that didn’t need it, water lost to bund leakage that goes unnoticed until the plot’s moisture is already off, and pest outbreaks that get caught only after they’re expensive to control.
What Smart Paddy Farming Actually Means
Smart paddy farming means putting sensors in the field to measure the conditions that drive yield — soil, weather and pests — and feeding that data into a dashboard farmers and extension officers can check in real time, instead of relying on schedules, habit and visual inspection alone.
In practice, this comes down to four things working together: a soil sensor and a tensiometer buried in the plot, a weather station on site, and a camera-based trap that uses AI to spot and count pests. None of these need to replace the farmer’s judgment — they just give that judgment something to work from.
Soil Sensors: Seeing What’s Underground
A soil sensor sitting in the paddy plot reads moisture, EC, pH and temperature continuously, and sends those readings back wirelessly so nobody has to walk out and check a probe by hand. The same data flags salinity build-up and pH drift early enough to do something about it — a dose of lime, an adjusted fertiliser rate — instead of finding out only once the crop has already responded to a soil condition nobody was tracking.
But moisture percentage on its own only tells you how much water is in the soil, not how hard the plant has to work to pull it out. That’s a gap a tensiometer closes.
Tensiometers: Knowing Exactly When to Re-Irrigate
A tensiometer measures soil water tension — how tightly the remaining water is held by the soil, expressed in kilopascals (kPa) — rather than how much water is physically present. Two plots can show the same moisture percentage and still be in very different states as far as the crop is concerned, because tension is what actually determines whether the roots can draw water out.
This matters most for alternate wetting and drying, or AWD, the irrigation practice most often recommended to cut paddy’s water use without cutting yield. AWD works by letting the field dry down between floods and re-irrigating once the soil reaches a set tension, commonly around 15 kPa measured at root depth, rather than on a fixed number of days. Without a tensiometer in the plot, that threshold is a guess. With one, it’s a reading — and re-irrigation happens exactly when the crop needs it, not a few days early out of caution or a few days late because nobody checked.
Installed alongside the soil sensor, the tensiometer turns AWD from a practice farmers are told to follow into one they can actually verify plot by plot.
Weather Stations: Field Data Instead of Regional Guesses
A general weather forecast covers a wide area and misses what’s actually happening over a specific scheme. A field-level weather station closes that gap — tracking rainfall, wind, temperature, humidity and light at the plot itself.
Cross-referencing rainfall against soil moisture and tension readings is one of the more useful tricks this unlocks: if a plot is drying out or losing moisture faster than rainfall and irrigation account for, that’s a sign of bund leakage or poor drainage, caught while it’s still cheap to fix. Wind data supports lodging-risk and spraying-timing decisions as harvest approaches, and humidity trends alongside temperature give an early read on disease and pest pressure — planthopper activity and blast disease risk both move with these conditions.
AI Pest Detection: Catching Outbreaks Before They’re Visible
A camera-based trap unit photographs the insects it catches, and an AI model identifies and counts them automatically — flagging species like brown planthopper, stem borer moth, leaf folder moth and rice bug before their numbers reach a damaging level. Instead of waiting for visible crop damage to confirm an outbreak, the count trend itself becomes the early warning.
Correlated against temperature, humidity and rainfall data from the weather station, pest counts also help flag which conditions are pushing risk up in the first place — turning pest management from a reactive spray-after-damage routine into something closer to a forecast.
The Bottom Line
None of this replaces the farmer or the extension officer — the agronomic decisions still belong to them. What it changes is what they’re deciding with. Instead of a fixed schedule, a general forecast and a visual inspection, they get continuous, field-level readings — moisture, tension, EC, pH, weather and pest counts — on the exact conditions that determine yield, cost and risk on that specific plot.
For a crop as central to Malaysia’s food security as paddy, that shift — from farming by habit to farming by data — is the difference between reacting to problems and catching them early enough that they never become problems at all.
Looking to bring soil, tensiometer, weather or pest monitoring into your paddy operation? Get in touch at info@autoflotechnology.com.