HomeBenefits of optimization applied to agriculture
.jpg)
The use of technology in agriculture has increased in recent years. The use of sensors to monitor soil properties; apps for remote monitoring; systems integrated to meteorological stations; and drones for marking the boundaries for planting and monitoring the crop and harvest, among others, are now a reality. These technologies are part of what has been known as Smart Agriculture and are the landmark of a new phase in agriculture - Agriculture 4.0.
The use of these technologies allows the capturing of a huge volume of data, allowing better decision-making. However, a huge database is of no service if it cannot be used wisely. And this is when the use of Data Analytics and Advanced Analytics techniques becomes extremely necessary and beneficial. Mathematical optimization, a branch of Advanced Analytics, has been used on several farms around the world and is a great facilitator in making better decisions.
Among the several decisions that can be optimized in agriculture, we can list:
- Amount of irrigation and fertilization to be made
- Dates of irrigations/fertilizations
- Amount of water to be used in each irrigation
- Choice of fertilizers
- Amount of each fertilizer
- Location of irrigators
- Location of agrochemicals applications
In general terms, mathematical optimization can be applied to any issue involving decision-making processes. For that, an objective must exist, so that the decisions involved in the model are made seeking to achieve that objective. There can be several objectives, and for agriculture, a few can be named, as follows:
- Maximization of profit
- Maximization of agricultural yield
- Maximization of nutrient absorption
- Minimization of costs
- Minimization of water consumption
- Minimization of consumption of fertilizers and/or agrochemicals
- Minimization of soil degradation
A single model can seek an optimal solution involving one or more decisions. Similarly, it can have a single or several objectives and have a short- or long-term vision. The restrictions may include chemical, regulatory, and operational constraints so that the final solution is consistent with reality and can be put into practice.
Depending on the chosen objective, society as a whole can benefit from an optimized decision-making. For the objective of minimizing the water consumption, for instance, it is possible to see the impact an optimization model can generate when we consider that agriculture represents 70% of the global consumption of water (FAO,2017) and that the world has approximately 750 million individuals without access to drinking water.
In addition, a survey developed by FAO (Food and Agriculture Organization of the United Nations) demonstrates that agricultural production must increase by 70% in comparison to 2009 to be able to feed the expected population of 9.1 billion people in 2050 (FAO, 2009). However, with a decreasing availability of natural resources, this problem becomes even more complex, emphasizing the importance of more efficient decision-making processes. In this sense, mathematical modeling is doubtlessly one of the possible paths for the necessary increase in productivity, allowing to produce more using the same space through optimized decisions.
Some people may ask how it could be possible to meet those objectives using an exact method, since some of the variables related to the agricultural process involve a great deal of randomness and cannot be predicted in an exact manner (agricultural yield itself could hardly be represented by an analytical function). In order to do that, there are several agricultural simulators in the market, which, from the input data, are able to predict the behavior of the soil and the growth of the crop. In this way, mathematical models can be used with those simulators. That type of solution is referred to as Black-box optimization or Simulation-optimization.
Whichever method is used, there are several impacts of the use of optimization on agricultural processes. They can generate benefits for the producers and society as a whole, whether by increasing agricultural yield, reducing costs, and/or reducing the negative impacts on the environment.
References:
al, J. e. (2003). The DSSAT cropping system model. European Journal of Agronomy.
Darshana, P. A. (2012). Simulation and Optimization for irrigation and crop planning. Irrig and Drain 61, 178–188.
FAO. (2009). Retrieved May 2, 2019
FAO. (2017). Retrieved May 2, 2019
Jones JW, A. J. (2017). Brief history of agricultural systems modeling. Agric Syst 155, 240–254.
Co-author: Emmanuel Marchal - Managing Partner at Cassotis Consulting