National Hog Farmer
Electronic sow feeding: An alternative that’s manageable

Electronic sow feeding: An alternative that’s manageable

ESF is an extremely flexible technology to implement and offers a variety of avenues to success.

Scientific evidence remains equivocal with regard to what is the best way to house gestating sows. However, both legislative initiatives and market forces will move sows out of gestation stalls over the next five to 10 years.

Producers who plan to build new sow facilities and/or stay in the business long enough to recapitalize their existing sow facilities will likely need to confront decisions about sow housing. Several alternatives to the gestation stall exist, but it is important that a producer’s choice matches his or her needs and abilities. The goal of this article is to share our experiences with implementing electronic sow feeding as an alternative to the gestation stall.

Experience with ESF
In 2001, an ESF system was implemented at the Swine Teaching and Research Center of the University of Pennsylvania School of Veterinary Medicine. The facility was built for teaching veterinary students and to provide a demonstration herd for alternatives to the gestation stall. By 2005, the basic model that was prototyped at the swine center was being implemented with our guidance on commercial herds. Today, our ESF system is feeding over 150,000 sows on 65 farms across the United States and Canada. The farms range in size from 100 to 10,000 sows, utilize a variety of common genetic suppliers, and are either family-owned and operated or company-owned and run with hired labor. The best farms are pushing above 30 pigs weaned per sow per year. Thus, ESF, if properly managed, is not a barrier to outstanding production.

Ways to implement ESF

There are several paths to the implementation of ESF based on the availability of existing facilities and/or access to ground for new construction. We divide these opportunities into three types of projects: new construction, expansion or renovation. The table summarizes the relative proportion of each type of project in which we have been involved.

Renovation is more common with smaller farms; expansion is embraced by larger farms, and new construction is divided more evenly between the two. New construction promises the best end result, as you have total control over the footprint of the barn and can design the gestation area without compromise.

This option, however, requires the availability of appropriately permitted land for construction. Expansion is most often used when an existing stalled barn is not totally depreciated or defunct and there is adjacent land available for building. The existing crates can be used as post-implantation stalled housing for sows early in gestation, and later in gestation animals are housed in pens located in a newly constructed addition. However, not all farms have the luxury of expanding their herd and a market for more piglets.

Finally, renovation or retrofitting is used when it is impossible to expand the footprint of the gestation building. It always requires some degree of compromise, as existing attributes of the building may not be easily altered and can impact the layout of the barn. Location of solid areas of flooring would be the most common example of compromise encountered when converting a partially slatted stall barn to ESF.

Slat-level considerations

ESF is an extremely flexible technology to implement and offers a variety of avenues to success. This does require producers to make several decisions during the planning stages in order to shape the farm to meet their needs. Here we review these opportunities and describe our experience with the different possibilities.

Parity segregation. We have observed good success on farms that practice some degree of parity segregation when organizing the flow of their animals through a pen gestation barn. Gilts are one of the most valuable assets on the farm, and the farms that have elected to flow the gilts separately from higher-parity sows are afforded several opportunities. These animals can be fed differently (e.g., a different ration than the sows) and managed differently (perhaps attended to by best available labor).

When mixed with higher-parity sows, the smaller, still growing gilt can be at a disadvantage and suffer negative consequences from being at the bottom of the social order. It may also be possible to segregate smaller P1 animals with gilts or to further subdivide the older parity sows (e.g., P2s) from the rest in an effort to reduce stress and competition in the pen environment.

Group structure

✓ Dynamic. In a dynamic system, group constituency is constantly changing. Essentially this is a continuous-flow system, and hence space utilization can be optimized. Dynamic flows are required on small farms where the weekly breeding group is less than that supported by a single ESF station (50 and 75 sows, depending upon the manufacturer of the ESF station).

We have used dynamic flows on 700-sow farms, on 1,400-sow farms that choose to flow gilts separately from the rest of the herd, and on 2,800-sow farms in combination with static pens. Dynamic flows work very well with big groups of sows as the social hierarchy is less rigid, so the addition and/or removal of animals from the pen is more easily accomplished. We have been able to take sows in and out of big, dynamic pens essentially at will and with impunity.

We typically target turning over the pen inventory by 10% to 15% with the introduction of each breeding group. Animals from multiple breeding groups are found in a given dynamic pen, and the animals compromising the weekly breeding group are divided and spread across several different pens. This creates an important challenge related to the implementation of a dynamic system due to the loss of the physical integrity of the breeding group. Management practices need to be adapted to accommodate the distributed nature of individual breeding groups, which often involves an increased reliance of management on the ESF system software and its ability to color mark or sort out specific animals within a pen.

✓ Static. In a static system, a group is constituted once social hierarchy stabilizes, and the group is left intact for the duration of gestation. The implementation of static groups is often attractive because the physical integrity of the breeding group is typically left intact, and some semblance of a breeding snake can be organized in the gestation barn. However, variations in weekly breeding targets and any unanticipated fallout from a group can lead to suboptimal space utilization of the facility as stable groups function as an all-in, all-out flow.

We usually implement static groups on large farms where a weekly breeding group is as large as or larger than the optimal number of sows supported by a single ESF station.

A 1,400-sow unit with a weekly breeding target of approximately 70 sows is the minimal-sized system facility for implementation of static groups. However, if one chooses to flow gilts separately from sows, as is preferred, this reduces the number of sows available in a week to constitute a static group to approximately 54 head. Depending upon the ESF station, this may be too small to fully utilize its capacity. We most often start the implementation of static groups with 2,800-sow barns. Here we would have about 90 sows and 30 gilts comprising the weekly breeding group.

We make one static group of roughly 75 sows per week, and then run the gilts and remaining sows in separate dynamic flows. This captures the convenience of a static system, but helps maximize space and feeder utilization as expected from a dynamic flow. Static flows are simplest with units with 5,600 or more sows. We more often see static flows coupled to post-implantation systems. On farms that are weaning and refilling farrowing rooms several days a week, they may constitute a static pen, but empty it dynamically — e.g., over a three- or four-day window. This can lead to some suboptimal space utilization in gestation, but it helps to maximize weaning age.

In sum, the decision about group structure largely depends on farm size. For farms practicing parity segregation of gilts in gestation, dynamic groups are used for herds of 1,200 sows or less, static groups for herds of 5,000 or more, and herds with size in between these can use some combination of static and dynamic groups to optimize animal flow and productivity.

Time of group formation

✓ Pre-implantation. Sows are crated after weaning and bred in the stalls. Groups are constituted as soon as animals are out of standing heat. The pre-implantation system also allows a sow earlier access to the ESF station during its production cycle, which has the potential to provide a more-sophisticated nutritional program than could be achieved in a gestation stall. The pre-implantation system minimizes the number of gestation stalls in the facility (e.g., typically two weeks of production or less) and prioritizes the management of sows in pens. However, a pre-implantation system may be less forgiving to implement as there is a three- to five-day critical window for moving of animals into pens that must be respected to ensure high farrowing rates and large litter sizes.

✓ Post-implantation. Sows are crated after weaning and bred. Groups are constituted only after being confirmed pregnant at approximately 35 days. Implantation is complete before mixing sows, minimizing the possible reproductive negative impact. This approach is initially attractive as it leaves the basic reproductive management of the sow unaltered from a crated barn, and the physical integrity of the breeding group can be left intact if farm size is large enough.

Our experience is that both pre- and post-implantation systems can support good production. Some of our best ESF herds are above 30 pigs weaned per sow per year and are using a pre-implantation system. Thus, pre-implantation group formation is also not unto itself a barrier to a highly productive sow herd.

Design of pens. While ESF is an outstanding way to feed sows, it per se does not do much to help mitigate the unwanted effects of animal-animal aggression common in groups of pigs. Successful implementation of ESF requires the management of social hierarchy in pen gestation. There are several important details in the design of the pen layout that help to ensure success. Some factors to consider are:

✓ Space allowance. Many of our farms were designed with pens having a stocking density of 18 to 20 square feet (1.67 to 1.86 square meters) per head. As producers have become more accepting of alternatives to gestation stalls and are more open to questions of how best to implement these alternatives, we have been putting some barns in at 22 square feet (2.04 square meters), which is similar to the European standard for a pen with more than 40 sows. As stocking density decreases (or space allowances increase), the social dynamics for the sows are less intense, and management of the pens is more forgiving. On new construction, larger space allowances translate into increased construction cost, whereas on a retrofit this can result in loss of inventory.

✓ Pen size. For dynamic pens, we favor the use of pens with two or three feeders per pen or at least 150 sows per pen so that the social hierarchy is less well maintained. That makes it easier to introduce new animals. As the number of animals in the pen increases, it becomes harder for barn staff to provide quality individual animal care. That’s why we typically do not use more than three feeders per pen, which results in a maximum number of animals in a pen up to approximately 230 head.

With static pens, we use a single feeder with 75 to 80 head per pen. In pre-implantation barns, we typically place 80 to 85 head in static pens to anticipate the standard fallout of a few sows at 21 days post-breeding. These recycling animals are moved from the pens to the breed row to be inseminated again, and space utilization in the pen is optimized.

✓ Pen shape. We most often use a rectangular-shaped pen. A single feeder is placed on the fence line along one side of the pen adjacent to the alleyway. The bedrooms are placed along the back of the pen facing the feeder. A natural traffic zone develops between the bedrooms and the feeder/fence line and allows sows to move unimpeded to and from the feeder. The rectangular shape of the pen ensures there is sufficient flight distance for a sow to escape her aggressor and increases the amount of perimeter for a given square foot of pen, as sows like to lie along the perimeter. All animals are also directly visible from the alleyway. For large dynamic pens requiring more than one feeder, we have connected the rectangular pens end-to-end along the alleyway. These large multi-feeder pens share all the attributes of single-feeder pens and allow for automated sorting of animals from the pen directly into the alleyway.The ESF technology is flexible in its implementation, so when necessary we have also utilized pens that are more square in their shape. In this case, the bedrooms are aligned facing at a right angle to the feeder. Appropriate spacing of the bedrooms allows for the development of similar traffic zones for sows to walk from their resting area to the feeder. However, sows are less visible from the alleyways in this layout. We have also applied a similar type of pen layout to large dynamic pens where multiple feeders are needed. Here the feeders are aggregated in a block at one end of the pen and may share a common sorting mechanism. The challenge with placing the feeders in close proximity is that a single boss sow has the potential to dominate multiple feeders. As such, we reduce the expected number of sows that can be feed by a single feeder in a day by about 10% to 15%, so this layout requires more ESF stations.

✓ Pen dividers. The use of pen dividers along the back wall of the pen will create “bedrooms” for the sows. This further increases the perimeter of the pen. That promotes orderly laying patterns and allows for the development of social subpopulations in the pen. We often observe the same animals lying in the same bedroom. We ask farmers to label each of these lying areas as that provides a convenient reference with which workers can find animals.

✓ Solid lying areas. We have farms that utilize either solid flooring or slatted flooring in the bedrooms. We are not legislated in our country with regard to flooring and requirements for solid areas. However, the animals will prefer the solid area for lying down compared to a slat; this helps to further order the pen.

✓ Waterer placement. Waterers are placed close to the entrance and the exit of the feeders. This discourages animals from sleeping there and creating congestion around the station.

The people factor

Successful implementation of ESF depends critically on careful consideration of the design and flow of the barn discussed above. However correctly addressing these factors does not guarantee success because people have the biggest impact on the outcome of the transition away from gestation stalls. People working in the barn must be enthusiastic and committed to the project, as well as equipped with the necessary training and/or skills to succeed. In our experience, the implementation of ESF usually goes easier on owner-operator farms as the person who made the decision to implement ESF is usually working in the barn and their personal investment is often at stake. This is in contrast to the situation in the more integrated production companies where the individuals who are tasked with making ESF work were not involved in the decision process, and they rarely have a significant financial stake in the outcome of the decision. Regardless of the size or type of the farm, it is important to identify existing staff members who are looking for new challenges and not resistant to change. Less pig experience and an on-the-job worker-training program can in some cases be preferable for at least part of the staff rather than trying to re-educate experienced workers from a crated barn.

In addition to selecting the right people to staff an ESF barn, providing these workers with the appropriate training is also critical to success. It is highly recommended that as many staff and for as long as possible visit/work in an existing ESF facility before the startup. The staff must learn the procedural differences that exist between a crated barn and an ESF barn. Common challenges include successful management of tasks new to an ESF barn such as electronic identification tag management (sows cannot eat without an EID tag, and this needs to be monitored daily) or training of gilts to use the ESF station (poorly trained animals will struggle to get sufficient feed intake in an ESF barn).

Another task that is dramatically different is individual animal care. The larger the pen, usually the harder it is to track individual animals. Staff needs to become adept at recognizing individual animals in need of care and then being able to repeatedly locate them. The former requires the staff to better understand sow behavior and movement in order to identify treatable problems at the earliest possible stage. The latter necessitates a system to mark individual animals or at least a labeling schema to identify specific locations of animals within the pens.

Finally, visiting a working ESF barn allows staff new to this approach to get a vision of how well such a barn can run as there are more ways to make the system not work than work. Staff will benefit from having a clear view of their road to success.

Upside potential of ESF

ESF offers several distinguishing features compared with other alternatives to the gestation stall. Our work has identified a variety of opportunities associated with ESF for the improvement of sow herd nutrition and management. This includes improved feed utilization by the reduction of feed needed to maintain individual animal body condition, better matching of feed delivered to changing nutritional needs of sows during gestation through the use of software-controlled feeding curves, automated control of gilt estrus via the delivery of Matrix to selected individual animals feeding in the ESF station, and a practical solution to regulating the amount and timing of a feedback program in pen gestation to stimulate immunization against autogenous pathogens. The electronic identification of the sows via a radio-frequency identification tag also opens the door for additional digital management of the herd such as spray marking of animals requiring vaccination or selection of animals to move to farrowing.

Based on these opportunities, we see ESF as perhaps the only alternative to the gestation stall that promises progress in how we will manage pregnant sows in the 21st century.


Our work has focused on the implementation of ESF on commercial farms in the U.S. and Canada. We have employed a variety of approaches to ESF including static versus dynamic pens and pre- versus post-implantation group formation. These choices were dictated by farm-specific details, and none were found to preclude outstanding production. Successful implementation of ESF cannot be captured without some forethought on how the barn will be staffed and how staff will be trained. In the end people have the biggest impact on the outcome. We see ESF as an important and readily implementable alternative to the gestation stall that offers a tool for producers to advance the management of their sow herds. 

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.