How Pigs Perceive Space

by 5m Editor
26 January 2004, at 12:00am

By Harold W. Gonyou, Prairie Swine Centre and W. Ray Stricklin, University of Maryland - Space allocation is an important part of barn design and management. When planning your facility you need to design your pens to accommodate the pig flow you expect. If you already have a barn, and litter size or market weights change, an understanding of space requirements is essential to create new pig flow patterns in your barn.

Visit the Prairie Swine Centre

There are three basic methods to studying space requirements for pigs: the empirical method, the behavioural space budget, and animat modelling. We have conducted studies involving each of these methods.

The Empirical Method (Study 1)

This method involves providing different amounts of space to groups of pigs and assessing production. The conventional method simulates commercial practice in that pigs are provided a set amount of floor area for the entire study. This usefulness of this method is limited as pigs are only crowded during the last few weeks of the study. The point at which the pigs become crowded is not known, and the effect of keeping pigs to a higher weight can only be speculated.

We used a variation of the conventional method which involves maintaining the same degree of crowding on the pigs throughout the study. To accomplish this we calculated space allowance on an allometric (body size) basis and adjusted the size of the pens every two weeks. Space allowance was calculated as a constant times body weight.667, which is proportional to the surface area of the pig. The constants used provided market weight pigs with normal levels of space, 25% more, and 25% less, as follows:

We also included 6 group sizes (3, 5, 6, 7, 10 and 15 pigs/pen) in a 3 x 6 factorial design, with two replicates per combination. Pigs were penned on a totally slatted floor, with negative ventilation, and temperature adjusted to achieve thermal comfort. The study lasted 12 weeks, at which point pigs were reaching market weight.

The most crowded pigs grew slower than those on either of the higher space allowances. The difference in performance among the treatments was similar during the first, middle and final thirds of the study:

The fact that crowding resulted in a similar level of reduction at all stages supports the use of the allometric formula for expressing space allowances. We can now confidently state that a coefficient of .039 will achieve near maximum growth of any size pig; at 30 , 60, 90 and 120 kg. If market weights increase, we can calculate the new space allowance for the larger pig.

Feed intake was similarly affected by space allowance, so feed efficiency was similar for all space treatments. Group size also affected growth rate and feed intake of pigs. In this case, production was decreased as group size increased:

We used a single-spaced feeder in each of the pens, but we do not believe that limited feeder space led to this reduction in gain with larger group sizes. We would not expect feeder access to limit growth for groups of 5-7 pigs, yet they performed poorly compared to groups of three. In the range of 10-15 pigs, when we might expect feeder space to be limiting, the reduction was not any greater than that observed at the smaller group sizes. We also made behavioural observations on these pigs and determined that the social orders in the larger groups were more complex than for small groups. We hypothesize that the complexity of the social order may affect performance.

Behavioural Space Budget

This approach involves studying the behaviour of the pigs and determining the amount of space needed for each behaviour. We know that all pigs will be lying during much of the day, and the space required for lying is approximately .027 * body weight.667. Between 5 and 10% of the time is spent eating, depending on feeder type, so an eating space is needed for every 10-20 pigs. A market weight pig takes up about .5 m2 while eating. A similar amount of space is needed for drinking, unless wet/dry feeders are used without an additional water source. We are less sure of the amount of space needed for dunging and general movement about the pen.

Studying the behaviour of pigs allows us to evaluate the ‘quality’ of space, not just the quantity. Quality refers to how well the space accommodates the pigs essential behaviours. Good quality space allows pigs to perform their necessary behaviours in less space. Our study examined the behaviour of pigs in pens with different configurations of feeder position and orientation. The pens were partially slatted with a nipple drinker in the middle of the back wall. A single-space dry feeder was used to feed the pelleted diet to the 10 pigs in each pen. After a preliminary study we narrowed our treatments to five configurations as shown in this figure:

Configurations 2, 3 and 5 were used in a production study and resulted in similar rates of gain and feed intake. During winter trials the pens remained relatively clean, with configuration 3 being slightly cleaner than the rest. The pens were dirtier during summer trials, with configuration 2 having the cleanest pens. Our general conclusion is that pigs generally keep their eating area clean, and by placing the eating area in or near the sleeping area, the pens are kept cleaner.

We studied the eating behaviour of the pigs in all five of the configurations. Pigs spent less time eating if their feeder was at a distance from the drinker (configurations 1 and 2). More time was spent eating if pigs could stand against the wall while eating from a feeder positioned along the side wall (configurations 3 and 5). The highest eating time was obtained when the eating space was outside the lying area (configuration 5). Feeders were sometimes blocked by pigs sleeping in the feeding space. Although none of these configurations affected intake or gain, the data indicates that configuration does affect behaviour, and under more competitive situations, such as larger group sizes, we would anticipate some effect on production traits.

The final portion of our study on pen configuration involved the use of an animat simulation. As described below, the animat program was used to estimate the distance pigs would travel in a day in order to eat, drink and sleep. Separation of the eating and drinking sites increased the distance pigs would have to travel.

Although this study on pen configuration did not yield a clear answer on the best way to organize a pen, it did suggest some features that should be studied further. Separation of the drinker and feeder affects movement, dunging patterns and time spent eating. The position of the feeding space relative to lying space also affects dunging patterns and time spent eating. Finally, having the eating space protected on one side by a wall affected eating time. The level at which any of these features would affect production is yet to be determined.

Animat Modelling

An animat is an artificial animal: either a computer simulation or a robot. We have used computer simulations in our studies. By using animats we are able to conduct a large number of preliminary studies to select treatments before conducting more expensive studies using pigs. To this point we have used what is known as a ‘top-down’ approach in our animat studies in which we control the complexity of the behaviour of the animats through programming.

Our animal studies have progressed through a number of stages. Initially our animats were randomly generated ‘points’ in a pen. This approach allowed us to determine some of the effects of pen and group size, and pen shape on the distance animals could separate from each other. The process was refined by giving the animats a 2-dimensional shape. The next step involved programming the animats to move, and the frequency that they ‘hit’ other animats was studied. We are now at a stage in which our animats move in response to their behavioural motivation. When an animat becomes hungry, it moves toward the feeder. When it is finished eating it moves to the drinker, and then to the sleeping area. The simulation includes several animats within the same ‘pen’, and they encounter each other as they move toward their goals.

The animats used in Study 2 were programmed to eat 12 meals per day, and spend 60 min/day eating. A diurnal pattern of motivation was used so that most meals occurred between 8 am and 4 pm. The pigs used in Study 1 were videotaped and their behaviour will be studied to improve the animat program.

A Current Study

Pigs are remarkably adaptable animals and our studies with relatively small pen and group sizes suggest that they can perform well in most configurations. However the industry uses larger group sizes which allow greater flexibility in pen design. We are currently conducting a study on group sizes of 20 and 40 pigs per pen, with four different pen configurations for the larger groups.

In a typical commercial barn, with feeders situated at one end of the pen, pigs sleep along the walls and make little use of the centre of the pen. Our alternative configurations, with feeders in the centre, corners or on each wall, may increase the use of the pen centre. The corner and wall configurations also allow pigs to form sub-groups within the pen associated with each feed source. In this study we are using single-space wet/dry feeders so that drinker position is not a confounding factor.


The most useful data we have for producers from our studies is the reduction of growth when somewhat less than .81 m2 (8.8 ft2) is provided for pigs up to an average pen weight of 95 kg. Because of the way we conducted the study, we feel confident that we can extrapolate the data to larger or smaller pigs. So if you market your first pigs at 105 kg, we recommend .81 m2 (maximum pen average of 95 kg). If you market at 120 kg, we recommend .90 m2. We may be able to lower these allowances as we obtain more data in the future.

We believe that the need for space is to accommodate the behaviours that pigs must perform in order to live and produce in their pens. Because of this we feel the best way of improving the quality of space, so that we can use the least amount of space without affecting production or welfare, is to study pig behaviour. We are studying the effects of several features of the pen which affect behaviour. These include the distance between the feeder and drinker, the orientation of the feeder along a wall, and the position of feeders in a large pen. The use of animats (artificial animals) is allowing us to reduce the time and cost involved in such studies.

Source - Prairie Swine Centre - December 2003