Kassim the game won by Eduardo Dolores Morales from Lima, Peru

Fun and educational game

Kassim is a tool developed by Wageningen University & Research (WUR) and has been used in higher education for several years. Kaspro, the high-end horticultural greenhouse climate and energy simulation model that drives Kassim, is used by teams in the Autonomous Greenhouse Challenge by teams to develop their control strategies.

While participating teams grow dwarf tomato plants in the research greenhouses of WUR in Bleiswijk, the Kassim game happened purely in simulation. The person who achieves the highest yield and the lowest costs wins. A fun way to learn about high-end greenhouse cultivation.

A shift in mind-set

“I saw it as a challenge”, explains Dolores Morales. “In addition to my experience and the knowledge I’ve gained through my professional career, my strategy to win was simple: understand exactly what inputs are necessary to achieve high yields with minimal costs."

“Instead of aiming to win, I focused on deeply understanding the game’s mathematical model. I wanted to test my hypotheses: if I input certain variables, would I get the expected output? This process helped me break free from paradigms shaped by where I’m from, (Lima, Peru), where greenhouse technology isn’t widely adopted, and agricultural practices are more pragmatic.”

“Another shift in mindset came from learning from competitors who scored higher than me. Observing them taught me that costs could be reduced even further, maximizing profitability. With all this knowledge, the final phase was trial and error—after all, it’s a game designed to teach through experimentation.”

Strong competition

Towards the final days of the competition, one competitor achieved a nearly unbeatable score of 540 points. Although my strategy remained the same, surpassing that score felt almost impossible. I spent several days and countless hours trying to break through, even reaching a point where I thought it might not happen. I was already set on agood plant density, stem number, fruits per trusses, thermal screen, avoiding LED lights and humidifiers, among other decisions to reduce costs. However, I realized that the most challenging parameters to optimize were temperature, CO₂, and the leaf area index (LAI). These factors are dynamic, changing with crop phenology and environmental conditions.

The winning strategy

“My final strategy focused on fine-tuning these parameters. For temperature, I started at 27°C to maximize growth, lowering it to 25°C for optimal fruiting. For LAI, I began at 3.5 to enhance photosynthetic rate and reduced it to 2.9 to balance optimal fruit development. Optimizing CO₂ levels proved tricky. Initially, I increased CO₂ gradually from 400 ppm to 900 ppm, but later I opted to start at 900 ppm and decrease to 740 ppm during fruiting, following the same logic: maximize early growth and stabilize during fruiting. This approach helped me achieve a score of 560 points, with potential for even higher—580 or 600 seems feasible with further fine-tuning.”

Besides eternal glory, the Eduardo Dolores Morales will receive a small token of appreciation.