Eliminating Human Error In Precision Ag Data Documentation
As precision ag consultants, we spend a lot of time working with clients to implement best management practices to acquire the most accurate data possible, and ensure proper machine and implement function. Probably one of the simplest ways we do this is to generate ‘Setup Files’ pre-populating displays with documentation information, including field names, machine offsets, and products. The main goal behind initiating this process is to eliminate human error when data collection occurs in the field.
There are functions within many of these precision displays which allow us to set a field locator function; ensuring data is acquired under the correct field name. Further, we now have functionality in many of our Farm Management Information Systems (FMIS), which allow for field detection and geospatial sorting of data to ensure it ends up in the correct place. With many implements such as planters, offsets and settings are saved in the implement controllers and automatically populated on our displays. These settings enable accurate function of processes such as section control on our planters.
The last piece of information that is needed for accurate documentation is the correct product and rate. This is one of the most challenging things to document properly. Part of the problem is that many products change from year to year, but the larger struggle is that the recording requires the most human interaction. Current precision ag technology has no way to verify the human input is correct. In our region of Ontario Canada, we still have a significant struggle in ensuring operators are accurately documenting their products. In order to document planted varieties correctly, equipment operators need to not only identify the correct variety in the displays, but also assign the variety to the correct planter rows. Things can get even more complex for sprayer operators when they need to document many products included in a tank mix. The constant struggle in agriculture to be timely adds another level of stress to acquiring proper documentation data.
It is very easy to see a world where agriculture moves to unmanned systems, and how documentation becomes not only a recording, but also a ‘work order’ for unmanned machinery. There are FMIS and mobile applications that exist today to create and generate work orders for operators. These work orders inform them of what equipment to use, location to perform the work, product and rate, and even what invoicing information to record at the end of the job. We could easily, instead of sending these work orders to human operators, send these work orders to unmanned machinery.
Currently, one missing piece and one which is difficult for me to comprehend why it is not more prevalent, is a barcoding system for our product documentation. Many of our products contain barcodes. Instead of leaving our documentation of products up to human input, we need a system in which we automate data entry in some fashion. Perhaps this is by a barcoding system, or maybe it becomes more complex where we have sensors that are able to determine active ingredients and at what concentrations or rates they are being applied.
Many of our FMIS are becoming systems that measure inventory, profitability, and break-even selling prices. It is difficult to quantify these parameters, when products are incorrectly documented or not documented at all.