Maximising pig data

Discover the importance of good recordkeeping and learn how to implement data capture proccesses to your business.

Accurate recording of data is essential and the increased use of mobile-data-capture devices and automated readers maximises accuracy and speed of data capture.

Recorded data is increasingly visualised for you; this means that examining the patterns and looking for the outliers (those events or animals that don’t fit the pattern), is easier and faster. It is the outliers that will lead producers to the processes that need to be streamlined to reduce cost of production or increase output. The most important thing is to always have targets for the business in mind and benchmark those targets to check progress.

Recording accurate data

The first step is to make sure that all staff are recording data accurately and the data is reflecting what is actually happening in the herd.

Stocktaking and correctly naming different groups of pigs is critical, otherwise, information can be skewed. For example, in a previous trial, a producer’s report was showing a high number of non-productive or ‘empty’ days. At first, this looked like sows were returning or aborting, but, on closer inspection, it turned out that the data set still included sows which had already been culled.

The use of handheld mobile devices allows farm staff to record data at the point the task is completed, registering births, weights, service, etc. This can all be done as the tasks are being completed. It also means the data is less likely to be lost, misread or left until someone else can enter it into the system. Increasingly, barcodes on semen packs, pen numbers and medicines are used to speed up events like AI, medicine use and movements.

Automated systems that use radio-frequency identification (RFID) tags in weighing and feed systems can automate tasks which were previously burdensome and time-consuming. It is important to remember to set the threshold low and keep it simple. As farm staff become more confident in gathering data with a mobile, it will become second nature, like sending a text while on the job.

Producers need to make sure the software groups their data in the same way as their pigs are grouped on the unit. For example, when a group of pigs is weaned, producers must also remember to tell the system how many have been moved or sold and where they have gone. This allows events throughout production to be linked. The software can inform the benefits of changing a task and the effect it can have on the rest of production.

Increasingly, producers are seeing the advantage of looking less at the averages of their herds and more at the variation. This is because the large numbers produced in the pig industry can hide the outliers that don’t meet specifications. In the UK, between 20–30% of slaughter pigs do not meet specification – depending on pig price, this is between 2p and 5p on every kg sold. To capture data accurately across individual pigs requires increasing amounts of automated and mobile data capture to ensure effective use of the farm staff’s time.

Running regular reports 

Increasingly, herd-management-software providers are developing new ‘visualisations’ to show at a glance how a stage of production is performing. These take various forms, from graphs and charts to colour-coded speedometers.

Image above: Average performance provides a snapshot of how a stage of production is working. The outliers shown in red and green numbers, show how much variation there is to allow producers to identify which production stages are most efficient and which are least.

Most importantly, these are moving away from average performance and starting to show the variation, the points in time where production has fallen out of ‘normal’ and how many outliers are in your production groups. These signposts are vital to make sure that the limited time farm staff have to spend on driving improvements is targeted, i.e. they can see from the report which are the most at-risk areas and fix those first.

Trials have shown that reviewing reports in isolation is less effective than using experts to help interpret existing reports and request new reports that are useful. It is also important to show or send the most relevant reports to everybody involved in the management of the unit so that causes and effects of any changes can be discussed and understood in more detail. Below are examples of the most useful reports to run:

1. Weekly performance against target (including targeted farrowings and numbers weaned)
2. Performance of sows by parity (to check gilt management policy, parity two dips, etc.)
3. When sows are returning (and whether they are being identified on time)
4. Effect of breed, age of gilt at first service and influence of lactation length on next performance
5. Parity distribution graph (to see whether the replacement strategy is working)
6. Stocktake (to check that the correct number and type of animals are in the recording programme)
7. The feed conversion ratio (FCR) and feed use
8. Medicine use and health status of stock
9. Slaughter data (to check number of animals out of spec and any deductions)

Talk to your software provider to check you are getting the most from the particular program and reviewing the most relevant information.

Observing the outliers of production

There is limited time and resource, constantly overhauling, changing and refining everything you do is not a realistic task. Visualisations that show were the outliers in production are sat make life easier, therefore it is important to seek those out in the reports generated by herd management systems.

For example, herd management software has a long history of reporting the average performance figures for numbers of piglets born alive, number of pigs weaned per sow and the typical mortality. More recently however, we are seeing reporting systems that provide not only the average figure but also then split down the herd to indicate the variation.

This simple display of those animals that over perform in a process and those that underperform allow farm staff to target the weakest areas first, making use of limited labour significantly more efficient. Some herd management software systems are going further and indicating those individual animals that have underperformed so that they can either be managed differently or removed from production.