Mixing & Grouping Strategies

Mixing aggression

One of the greatest concerns with group housing is aggression among sows. Injuries sustained during fighting are a significant concern for producers, as severe injuries can result in the need to cull valuable animals from the herd. However, research on aggression in sows has found the majority of injuries sustained in group sow housing are superficial scratches around the neck and shoulders, and less than 1% of the injuries observed are considered severe.  Overall, sows in groups generally ‘get along’ and there are many management techniques to reduce the incidence and severity of aggression. To effectively reduce and manage sow aggression, it is important to understand what causes it in the first place.

Some aggression will naturally occur at group formation as sows establish their social status.  Aggression at group formation can be intense, but is short lived with most fights occurring during the first 2 hours after mixing. Fighting can also take place during daily feeding, especially in competitive feeding systems. Daily aggression over feed access is a more serious issue, causing chronic stress and potentially the removal of poor-doing sows from the group.

In the wild, groups of sows include related individuals and their offspring with each group having its own territory and social order, and will drive away unfamiliar animals. When penned together on-farm, sows do not have the option to leave and the animals will fight to establish their dominance within the group. The initial mixing aggression is short lived and intense, typically involving the more dominant and unfamiliar animals. Afterwards, social tolerance develops as sows communicate using more subtle ‘avoidance’ behaviors. The tolerance of new group members develops gradually in a process lasting several weeks. Thus newly introduced animals will often be seen lying away from the main group of sows, often in less preferred areas of the pen.

Techniques for reducing mixing aggression include grouping familiar sows (when possible), providing sufficient floor space per sow, and the use of pen dividers (partitions) to give sows divided space and hiding areas to escape aggression. Pen partitions are typically seven feet in length, and help to define lying areas as sows prefer to lie against a solid wall. While partitions are typically four feet high, studies completed in Denmark indicate that short walls (e.g, two foot) can also be effective. Short walls can improve air flow and make it easier to see sows in the pen, but care should be taken to ensure that they are not placed in areas where sows may jump over them causing injury. Large, rectangular pens will also allow more room for sows to avoid one-another and good flooring (eg. partially slatted, gaps no greater than 0.8 inches) will reduce the likelihood of hoof and leg injuries.

Static versus dynamic groups

Gestating sows can be managed in either static or dynamic groups. For static groups, all sows in a group are mixed on the same day, and remain in that group for the rest of gestation with no new animals added. Static groups are used in competitive feeding systems, and in some non-competitive systems. In competitive feeding, sows groups should be formed of sows uniform in size, as this helps to even out competition. With only one mixing at group formation, static grouping allows sows to form a stable group hierarchy, and helps to reduce competition and allow all sows to feed. In competitively fed pens, sows should be observed daily at feeding and any that are injured or losing body condition should be removed promptly to comfort pens. No new sows can be added to small static groups, as any new animals will be targeted for aggression from other sows.

In dynamic groups, small groups of sows are added to a larger existing group periodically throughout gestation, and groups are also removed periodically as sows move to farrowing. Each time a new group is added a new bout of aggression will occur. However, with large dynamic groups it has been shown that sows adopt a more tolerant and passive response to unfamiliar animals. A pen layout that includes adequate space, and the addition of dividing walls for escape and to hide behind, helps to reduce the aggression.

Group size

Group size can also influence sow aggression within the group. Smaller groups of sows (8-10) will form a social structure with stable, linear hierarchies. Larger groups of sows (≥40 animals) develop a different social structure in which there is a greater tolerance and less aggression. It is believed that the cost of establishing dominance in a larger group is greater, and so animals learn to adopt a more tolerant approach rather than trying to dominate. Instead, sows in large groups will form smaller sub-groups, or cliques of sows that lie together.

With dynamic grouping, maintaining large groups is one way to help control aggression. In the dynamic setting, the addition of a smaller group of sows to the larger group lends itself to the natural formation of sub-groups of sows. To help reduce the stress to animals added to a dynamic group, it is recommended that animals added to a new group should make up at least 10% of the total group size.

Many producers use an introduction pen that allows incoming sows to be penned together and become familiar with one-another before joining the larger group. These sows typically remain together as a sub-group once they join the larger group. A specially designed mixing pen can also offer better conditions to minimize aggression and stress, including features such as extra space per sow, better flooring to help prevent injuries and provision of enrichment or bedding.

Grouping by size

Grouping sows by size (and/or parity) can help to even out competition within the group, helping in particular the smaller, younger parity sows to do better by not having to compete against larger, older sows. Grouping by size is especially important in competitive feeding systems where smaller sows will be at a distinct disadvantage when competing for feed.

Research has also shown advantages to grouping sows by parity in non-competitive feeding systems. Grouping sows by parity was found to be advantageous for parity 1 sows managed in a static ESF system. Younger sows in the study were able to maintain backfat when housed together, but lost backfat during gestation when kept in mixed parity groups. It is likely that maintaining the sows in uniform parity groupings reduces competition for entry to the ESF feeder, helping to ensure all sows receive their daily feed allowance. Research indicates that grouping sows by parity can also benefit young sows fed in free-access stalls. When housed in mixed parity groups, younger sows spent the majority of time in the free-access stalls, but when housed with other young animals the sows spent more time out of stalls in the common loafing area.

Moffet Farms. One of the bedroom sleeping areas full of sows. Water drinkers are on the outside of the bedroom pens.
Sows grouped by size in gestation pen sleeping mid afternoon

Research: Mixing times for sows

Currently, the most common management strategy for group housing in North America is to place sows in groups at 4- 5 weeks after breeding. This avoids aggression during the implantation period, allows for individual feeding and observations during breeding and implantation, and allows pregnancy checking to be done in stalls. However, further reductions on stall use are possible and other mixing times may offer some different advantages. For example, mixing directly after weaning would reduce the amount of time sows spend in stalls, increase the amount of space available for gestation pens, and would allow the majority of aggressive encounters to occur before breeding.

The Prairie Swine Centre (PSC) recently concluded a study looking at different timing for mixing of group housed sows. The strategies tested included early mixing (EM: mixing at weaning), late mixing (LM: mixing at 5 weeks), and pre- socialization (PS: mixed 2 days, then stalled until mixing at 5 weeks after insemination). The pre-socialization treatment was tested as a way to reduce aggression when pregnant sows were regrouped. Each treatment used groups of 14 sows, fed in free-access stalls, with sows allowed into stalls only for feeding.

The results showed that there was little difference in production figures or the level of aggression among the different strategies. The conception rate was highest for early mixing, and lowest for the late mixing treatment. Early mixed sows also had fewer still- borns when compared to late mixing or pre-socialization (Table 1), which could reflect the increased fitness of sows from the additional time spent in groups. There were no significant differences in aggression between treat-and the pre-socialized sows experienced aggression at both mixing times, so it did not show any advantage over the other treatments


Variable EM PS LM P
Conception rate (%) 97.62 94.05 86.9 0.028
Total born 15.16 15.63 15.47 0.700
Born Alive 13.66 13.27 13.18 0.691
Still born 0.95a 1.54b 1.58b 0.003
Mummies 0.47 0.44 0.53 0.766

Table 1. Production characteristics of sows in three mixing treatments: Early Mixing (EM); Pre-Socialization (PS); and Late Mixing (LM).

Overall, the production figures among treatments were similar and indicate that the timing of group formation is flexible. Mixing at weaning can allow producers to reduce the amount of space taken up by stalls, and may be preferred for those wishing to decrease stall use. Mixing in the standard way at 5 weeks after breeding can allow for reduce labour at breeding, heat checks, pregnancy checks, daily health checks, and individualized feeding throughout the implantation period. Pre-socialization is not recommended due to sows experiencing initial mixing aggression twice, and the increased labour for stockpersons. Further studies to examine the effects of mixing after insemination are now underway at both PSC and the University of Manitoba.

The mixing at weaning study received international recognition as Dr. Jennifer Brown, research scientist at Prairie Swine Centre, was presented with an innovation award from the U.S. National Pork Board at the American Society of Animal Science Midwest meeting in Des Moines Iowa, in March 2015. The research was a collaboration between the Prairie Swine Centre and the University of Minnesota, and was supported by funding from the National Pork Board