Digitally assisted agricultural machinery and equipment can already be found in arable farming. A few examples of this are self-steering tractors, application technology with variable dose rates for fertilizers and plant protection agents, and automatic yield logging with combine harvesters. Milking robots, sensors, databases and various items of digital auxiliary equipment have been assisting farmers in animal husbandry for a long time too.
Digital farming goes beyond this and stands for complete internal and external IT networking of previously isolated individual systems on a farm. In the networked state, the interaction between machinery and production processes goes well beyond the ISOBUS standard. This leads to the development of altogether very complex production systems, made possible by mobile telecommunications and internet-based portals. Such systems are very promising, as they display great potential for advancing a wide range of optimizations in agriculture.
However, this does not mean that with the digital options everything will proceed completely automatically as on a factory floor. Agriculture is not the same as industry. This is due to the fact that in industrial production the same production conditions prevail in buildings and halls on every day of the year. Farmers on the other hand work under open-air conditions and are highly dependent on the weather. Changing, interacting and random variables are the hallmark of the agricultural production process. That is why in future too farmers with their experience will remain indispensable. They will need to intervene and take corrective action or decide between variants that might be offered by a digital system.
What we understand today by Precision Farming is chiefly site-specific/subplot machinery and equipment systems and automation in general. Site-specific treatment systems abandon uniform dose rates and adapt doses to the changing conditions within the plot. In automation, substantial savings of farm inputs, energy and working time are being achieved for example via auto-steering systems and boom section controls. Smart Farming has extended Precision Farming by greater use of real-time sensor technology, including data fusion for decision-making support.
In digital farming we are at the start of developments now. This is where new components such as machine-to-machine (M2M) communication (Internet of Things), cloud computing and Big Data techniques will be used in order to make use of existing potential for optimising complex agricultural production systems.
Today site-specific subplot treatment frequently fails due to the fact that with the Precision Farming methods available so far, the volume of information can no longer be handled manually. Furthermore, for example the fertilizer dose rates are modified based on just one parameter, although frequently several parameters need to be taken into account. However, it will be possible to avoid these deficits in future when map-based and sensor-based systems are merged closer together and work automated in real time. Then ever more relevant parameters – whether mapped or actually measured – will be taken into account. This could be a new solution helping the site-specific technology of Precision Farming to achieve its breakthrough.
This also applies for the entire farming operation, because a comprehensive system analysis can enhance the sustainability of overall production. This conserves resources and ensures that environmental protection rules are observed, for example because the field sprayer or fertilizer spreader switch off automatically when they come within the minimum distance from waterways. The system analysis and transparency in combination with an electronic yield map allow uninterrupted documentation of production, which provides the farmer with major benefits in many aspects.
Special Big Data methods can be put to expedient use in cross-farm application. Here regional data pools make it possible to analyse regionally characterised information such as stock management, variety behaviour, use of active ingredients, or timing of applications. This analysis could help the individual farm manager enormously, as here it is not just the experience of the individual which is directly available, but instead the experience of many colleagues with similar problems from the same region. Farmers from the same region could release and use information for certain crops jointly on a platform. Other service providers could, at the wish of the farmers, analyse the data comprehensively and recommend the resulting optimized measures on a site-specific or even subplot-specific basis here.
However, digitization of agriculture does not represent a new stage of mechanization. This means that the costs, for example, are dependent less on the utilization rate of a machine, but instead are more process-oriented. It is to be expected that small and medium-sized farms too will be able to afford this technology, as it frequently works with existing hardware and therefore has a more structure-preserving effect as regards the farm size.
Digital farming can generally benefit organic farming as well, as this is where up-to-date knowledge of stock development and field conditions make better production results possible. Close combination of organic farming and robotics would also be conceivable, as autonomous machines open up new possibilities of promoting biological intensification altogether with greater resource efficiency and biodiversity. They can loosen the soil as required by means of sensor technology, place seed in the soil extremely gently and evenly, care for crop plants during vegetation and remove yield-relevant and quality-relevant vegetation. Site-specific technology is expedient here for soil working, drilling and nutrient application as well. Farm inputs will be reduced to a minimum and both high yields and qualities as well as high sustainability and environmental conservation will be achieved.
When digitally networked systems are used in agriculture, high priority should be accorded to data protection. The site-specific farm and business data may only be used if the farmer authorizes such use, as data from agriculture have now become tradable goods of significant economic value. Where business models are used, the farmer as owner of the data must benefit economically from such trade and the business operations must be documented transparently.
At present sensitivity with regard to data protection rules is growing among those politically responsible, as in future the safety of food production will depend more strongly on digitally networked systems.
This makes hacker and cyber attacks dangerous. The fact that the German Federal Office for Information Security (abbreviated to BSI) reckons agriculture among the so-called “critical infrastructures” underlines the relevance of this issue.