Yield-potential graphs for hydroponic greenhouse tomato production were developed for a decision model (HYTOMOD). In all cases, yield represented marketable fruit and the highest yield was assumed to be a function of both harvested weight per unit area and high quality fruit.
The model and curves are unique in that they are based on extensive scientific literature and were tested and validated by four independent expert growers. During the validation process, experts actually found the competition with HYTOMOD to be stimulating to their own skills and knowledge!
While the yield-potential graphs are an essential component of a computer decision model, they can also be used effectively as teaching aids. Each graph can be useful to new growers for basic plant production guidelines and each is a unique tool for experienced growers to refine their growing skills. These graphs are accessible at this site in the Graphical Primer. The program HYTOMOD has been rewritten in Java so that it is interactive and accessible over the internet. It is also available here in the Interactive Decision Support section.
Successful hydroponic tomato growing is assumed to be based on five key tests during each of five growth stages. The five growth stages are:
- germination and early growth
- early fruiting
- mature fruiting
The five key tests are:
- Hydrogen ion concentration
- Electrical conductivity of the feeding solution
- Root temperature
- Greenhouse air temperature
- Relative humidity
Night and daytime light levels are also independent considerations for all growth stages except germination.
This decision support system, implemented in the Decision Support Algorithm, is a model based in utility theory. The optimal growing conditions for tomato plants are known to the algorithm, and they are stored in a database. These optimal growing conditions are known to be acceptable because they are obtained from peer reviewed scientific literature. The model, is a model of the behavior of a hydroponic tomato grower. The model can recognize when growing conditions are not optimal and recommend which conditions need to be modified. It can also prioritize them according to the manager’s preference, this preference is called a utility. These utility levels allow the conditions of the crop to be assessed. These values are not the same as risk because they don’t predict whether a crop will be successful or not, they communicate the typical expert grower’s preference. This preference will in turn be related to risk and other factors. Another factor which affects the utility level is the managers aspiration level. This software driven manager aspires to having a high yield, high quality crop.
Special thanks for this effort goes to a hard working engineering graduate student from Egypt, Ahmed El-Attal for his work in creating the original version of the model (HYTODMOD). Also, our thanks to Jim Brown of CropKing, Inc, Seville, OH. Mr. Brown’s intensive course for beginning growers gave us the focus for this work