Introduction

Pork producers and consumers are concerned about any impact of production environments and management practices on pig well-being and performance. The productivity and welfare of the pig has major influences on sustainability and profitability of a pork production operation. Effects of production systems on the pig is a major public issue, driven primarily by public perceptions and regulations promulgated in Europe, not by science. Ironically, forcing producers to make unwise changes in their production systems could unwittingly compromise pig welfare. It is important that producers have at hand scientifically developed optimal production schemes and systems that will enable the industry to improve the health, productivity, and reproductive efficiency of the individual pig while sacrificing neither well-being nor profitability. We must gain information on animal well-being and economic implications of alternative production systems if we are to be in a position to make decisions that will enable us to optimize behavior- based production technology, such as auto-sort, in terms of animal health, performance, and well-being, while maximizing profitability.

Scientific literature reporting any impact of an auto-sort system on pig well-being is limited. Largely anecdotal trade-press reports provide claims of substantial benefits of auto-sort technology at loadout time in many operations and possible welfare benefits, but there is little information to help producers understand how to implement such a system or to decide which of the currently available new construction or retrofit layouts to use. In fact, many of these decisions are being driven by the companies designing the equipment and systems. Consequently, auto-sort is a relatively new but fast-growing, behavior-based production system that is being implemented by early adopters without a full understanding of how these systems should be used in order to gain any of their touted benefits. The only way the industry can be successful in adopting this technology for either new-construction or retrofit situations will be by generating information based on scientific evidence that will facilitate improvement of designs, layouts, and husbandry schemes.

Alleged auto-sort advantages

In summary, the trade-press reports that exist have implied that the auto-sort floor layouts have economical benefits due to reduced sort loss and less labor. A study conducted by Brumsted and colleagues (2004) reported that during transport fewer death losses were measured among pigs that were managed in an automatically sorted barn (0.11 %) compared to those managed in a conventional finish environment (0.21%), thus implying that pigs from large automatically sorted barns are more conditioned to walk, and thereby making loading and trucking less stressful. Others have claimed that these systems provide an opportunity to increase the throughput and sell more hogs at the top weight range. A 1,000-head producer can potentially gain $7,000-10,000 per year (Kelley, 2004). A speaker at the Iowa Pork Producer Meeting in February of 2005 indicated that the autosort system improves animal welfare because it enables pigs to move around, smaller pigs are less likely to be “picked on” because they get lost in the crowd, pigs can choose their “comfort zones,” and, finally, pigs load better at market time. Also, there have been claims that the auto-sort system reduces tail biting, fighting, and death loss; thus welfare is improved. Moreover, these anecdotal trade-press reports conclude – from an economical standpoint – that there is better uniformity in size of pigs produced, accurate marketing is enhanced, labor savings realized, and improved feeding and throughput achieved.

The economic benefits of an automatic-sorting system are thought to be mostly due to labor savings, easier feed withdrawal, reductions in sort variation and sort loss, greater uniformity in pig market weight, and therefore more accurate marketing (Vansickle, 2004). Sort loss reportedly has been cut in half in auto-sort barns as against conventional finishing facilities (Vansickle, 2004). It has been estimated that an auto-sort barn can save a producer about $2,000/1,000 head due to heavier pigs being sold prior to closeout and $250/1,000 head in sorting labor cost (Connor and Lowe, 2002). Others have reported that accurate sorting and marketing decisions can be realized by the auto-sort, as well as marketing more hogs in the top weight range; thus a producer could potentially gain $27 a day or $10,000 annually (Li et al., 2003). Moreover, the auto-sort system provides the opportunity to sort pigs into different pens for different diets based on body weight or on sex. Data from Kansas State University indicate that sorting pigs by weight and feeding a higher energy diet to the lightest 50% of the pigs increases average daily gain and reduces the number of lightweight pigs sold (Hastad et al., 2005).

Some have proposed that the auto-sort system improves animal well-being because it enables pigs to move around and find their individual comfortable temperature niche or flee aversive situations and get lost in the crowd (reported by H. Gonyou, Vansickle, 2004). Pigs in large-pen auto-sort systems seem to be less leery of people than those in conventional pens. Moreover, there have been claims that the auto-sort approach reduces tail biting, fighting, and death loss, thereby improving animal well-being. Brumsted and colleagues (2004) reported that, during transport, twice as many pigs that were finished in a conventional environment died as did those finished in an automatically-sorted finish system. Finally, there is general consensus that pigs from auto-sort systems are easier than conventional pigs to handle, especially during truck-loading and -unloading. Pigs with autosort experience are accustomed to walking single-file through chutes and gates and move easier into sort pens as well (Gonyou, 2005). It takes fewer people to load-out an auto-sort barn, and the pigs load faster and with less effort.

Several largely anecdotal trade-press articles allege substantial benefits of auto-sort technology at load-out time, and this technology is being adopted based mainly on those reports. The reality is that some auto-sort systems have been successful, but many more have been unsuccessful; thus, there remain a lot of questions that need to be answered before automatic pig-sorting systems can be unreservedly recommended to the pork industry as a production-system component.

Alleged auto-sort drawbacks

However, despite these claims, there appear to be drawbacks associated with automatic-sorting systems: (1) a learning curve comes with owning an auto-sort system – effective training of pigs and top management are essential; (2) the system can increase the overall cost of your building; (3) durability issues; and (4) limited research data exist at this point (Vansickle, 2003). But, more importantly, no scientific information exists about these systems to guide producers in terms of pig welfare and performance or operation sustainability at times other than preparation for load-out. It is imperative that producers understand the biologic and well-being implications of the auto-sort system, because if a pork-production system is biologically infeasible, then concerns about its economic and wellbeing impacts will be moot.

Other factors’ influence

There are many factors – including (but not limited to) equipment design, environment, genetics, management, nutrition, stressors, and so forth – that may influence the overall behavior, performance, health, and productivity of pigs grown in these various autosort systems. In fact, one of the biggest factors that may influence the success or failure of these auto-sort systems is social stress. Social stress undoubtedly has an impact on pig performance and well-being. This impact has been partially quantified and modeled by Wellock et al. (2003, 2004). Their findings imply that a decrease in initial body weight variation and an improvement in pig adaptability (coping) may be a better way to improve overall pig performance than selecting only for increased growth potential (Wellock et al., 2004). Unfortunately, social stress is generally ignored as a potential factor influencing the pig’s performance potential. Thus, it is essential to understand the most important factors that contribute to performance, productivity, and well-being of pigs being managed in these auto-sort systems, so that these systems can be optimized, regardless of the genetics, management, environment, and so on. They must work for everyone.

Questions

Currently, interest in auto-sort technology is being driven primarily by the marketing activities of firms manufacturing and selling this kind of equipment. As a consequence of these activities, producer interest in knowing more about this new management approach has been stimulated. But right now, producers are in need of answers to several questions.

1. Does use of an auto-sort system throughout the grow-finish phase affect pig growth rate, variation in body-weight gain, feed-conversion efficiency, carcass quality, health, or state of being? If so, then how?
2. Does use of an auto-sort system lead to reduction in the numbers of downer and dead pigs during transport and at the packing plant?
3. Does use of an auto-sort system reduce variation in sort-loss at the packing plant?
4. Is implementation of auto-sort technology economically feasible overall?

To answer these questions, the most important factors affecting pig performance and well-being in an autosort system must be scientifically identified and characterized. To date, they have not been identified, let alone characterized.

Current status of auto-sort technology:

Producers’ and firms’ assessment

There are auto-sort systems in operation that have been successful, and those producers have gained the benefits associated with this technology. Some of their testimonials have been reported in trade-press; others have been used by manufacturers as they market their systems. More recent data acquired from a producer that uses an auto-sort barn layout and the "water pen" in conjunction with phase-feeding indicates an improvement in average daily gain, feed to gain, and throughput by 7 to 10 days compared to their conventional small-pen barns (personal communications). This 1,000-head wean-to-finish auto-sort barn also resulted in a feed savings of 38.6 to 48.3 lbs per head. This particular auto-sort layout provides the opportunities for the pigs on the lower end of the body-weight distribution to benefit. Despite the positive benefits gained from this particular auto-sort layout, it is important to emphasize that this layout and management scheme is more complete than simply installing scale units and gates.

Despite this report of success, there are many producers who have adopted large-pen auto-sort technology only to experience unsuccessful results. Some adopters of auto-sort have yet to regain initial capital outlays; others have become frustrated and have removed the equipment from their buildings.

Current status of auto-sort technology:

Some scientists’ assessment

From a scientific standpoint, at this time the concept of auto-sort technology is still at the theoretical stage. What is known from ample empirical evidence of which we are aware is that this technology is by far more complex than simply installing an automated-sorting scale unit and a set of one-way gates in a hog house. This complexity is evidenced by what appears to be incomplete animal adaptation, various design flaws, and unintended and unexpected consequences of these systems. These problems – among others – have been confirmed by observations both casual and controlled over the past 18 months by a research team at the University of Illinois working on a project funded by the Illinois Pork Producers Association and National Pork Board.

An on-going research study was designed to evaluate the impact that current auto-sort floor layouts have on pig performance and well-being and to further optimize auto-sort layouts based on pig behavior. Currently, experimental auto-sort layouts have been assembled at two separate hog units in Illinois in which we are examining the design and operation of specific components of auto-sort technology: the scale unit, its entries and exits; the one-way gates; the various layouts; location of important resources such as feeders and waterers in each of the various layouts now being employed; novel layout approaches; relative plan dimensions of different zones in respective layouts; and so on. Many thousands of observations and measurements of various kinds already have been made. Video-records and ear-tag-RFID weight and scale-use data from many hundreds of hours of observations of thousands of pigs have been captured. Packer and Pig Champ data have been collected as well.

Through three replicates of 600 pigs per treatment group, careful observations and data analysis has confirmed that many flaws do indeed exist and benefits are there, but not yet optimized. Behavioral vices and patterns are significantly impacted by the various auto-sort floor layouts (food-court, water-pen, and fast-lane). These floor layouts differ in the location of resources as well as the entrance into and exit out of the system. Early observations revealed that some equipment (beg, scale and gating) per se and some layouts encourage and provide opportunity for "bad habits" to develop and become engrained if pigs are not adequately trained, and/or equipment malfunctions. For example, in one wean-to-finish operation, it has been observed that pigs will loiter in and around the equipment and use the scale unit as a place to escape the crowd and to perform stereotypic (repetitive, nervous, exploratory) behavior. They also will use gate entryways to guard other pigs from exiting a zone. We speculate that this problem becomes intensified due to inadequate training of pigs, malfunctioning of equipment, and equipment design. Also, oral-nasal-facial behaviors were significantly increased in this operation; thus, when a specific animal acts out these behaviors, a slow-down develops in the system such that pig flow is virtually halted. Scores of pigs were prevented from moving to the zone they wanted to be in, and the traffic jams and back-ups can become enormous, leading to frustration and stress.

More recently, under normal circumstances (independent of inadequate training or malfunctioning equipment), oral-nasal-facial and aggressive behaviors were observed and found to be more prevalent in some auto-sort layouts compared with others. In fact, the majority of the aggressive behavior was observed early in the training phase as well as in the later finishing phase. The aggression observed in the finishing phase was mainly around feeders and waterers; thus it appeared that these resources were restricted. Other important behaviors (ie, eating, drinking), behavioral sequences, and overall activity are all affected by these various auto-sort layouts as compared to large conventional pens. The alteration in behaviors and behavioral patterns appear to be due mainly to the way pigs access their resources and the location of these various resources that are driven by the floor layout of these different auto-sort systems.

Several other conspicuous problems have been observed as well. Scale hits are not correlated with growth rate. Sort-loss occurs at a higher rate in some of the auto-sort layouts than others. For example data obtained for a third and final group of pigs from an auto-sort system at load-out showed a financial sort-loss to the producer not only due to light pigs but heavies, too. In fact, three heavies cost the producer $14.00 per cwt. Pigs may simply avoid the sorting processes till the very end of their time in the barn. Across all three replicates, on average, sort-loss was least for pigs from a food-court layout compared with all other treatments.

Despite data depicting design flaws with the current auto-sort layouts in this experiment, data collected thus far does reveal some positive attributes of auto-sort technology. In general, the majority of pigs within a group learn rather quickly how to use the auto-sort system regardless of the floor layout. For example, the time to exit the scale on average decreases from 55 seconds per pig on day 1 of training to < 10 seconds per pig by day 2.

From a performance standpoint, in general, feed cost per pig was similar among pigs from auto-sort systems and large conventional pens, with the exception of one auto-sort layout (food-court); feed cost per pig was slightly less. However, on average, feed conversion ratio was similar among all treatment groups. Pig average daily gain was similar among all auto-sort treatment groups. Percentage of lean was consistently higher among pigs from auto-sort systems than from large conventional pens.

Pigs from these auto-sort systems handle better and load faster than do pigs from large conventional pens. and within the different auto-sort treatments there were differences in handling scores and loading times as well. The use of prods during handling and loading was less among pigs from auto-sort systems compared with pigs from large conventional pens. Also, percentage of vocalizations observed was ~50% less among pigs from auto-sort systems than conventional pens. In fact, pigs in auto-sort treatments appear to be less flighty and fearful of people than those pigs in large conventional pens. Pigs from auto-sort systems move through the packing plant better than those from conventional pens. Finally, there appears to be a reduction in the number of “dead” and “downer” pigs observed at the packing plant from these auto-sort systems.

Summary

At this point in the middle of 2007, we think it would be unwise to make any concrete recommendations on adoption of auto-sort technology in view of the current lack of understanding of the pig-equipment interfaces in the auto-sort systems currently available on the market. Many factors still need to be addressed and many improvements still need to be made. Based on our early observations and more recent data collection, we can say with confidence, at least, that in order for auto-sort systems to operate in ways useful to the industry, the designs of these systems first must be optimized based on the nature of the pig (especially pig behavior) while maximizing pig flow through the system without restricting access to critical resources, such as feeders, waterers, and comfortable resting and dunging places. Can the current problems observed and recorded, and expressed by users of auto-sort systems and our research team be overcome? Probably. Moreover, some of the proposed advantages of auto-sort technology probably can be realized by producers in a cost-practical scenario. But in order for auto-sort technology to be beneficial, practical, and profitable, we must continue to troubleshoot problems and answer the most pertinent questions about this novel concept with science-based research. Obtaining the necessary information about this promising technology will enable us to optimize the design of auto-sort systems while providing choices so that producers who want to consider adopting this technology can do so regardless of the diversity of unique circumstances at their respective operations. Auto-sort systems, at minimum, should facilitate the natural behavior of the pig, enhance its overall well-being, increase its productivity, improve its product, and provide the producer with adequate returns on investments of capital and management. Those goals are not now being consistently achieved. Acknowledgements The author gratefully acknowledges the University of Illinois research team, various collaborators and producers for their support and effort of this project. The author also gratefully acknowledges the National Pork Board and the Illinois Pork Producers Association for their financial support of this project. This paper was previously published in the 2007 AD Leman Conference Proceedings.

References

Brumsted, M. 2004. Automatically sorted, large-pen finish barns reduce death loss during transportation to market. Proceedings, American Association of Swine Veterinarians (AASV) Annual Meeting. Perry, IA: AASV. Online: http://www.aasp.org/

Connor J.F., and J. F. Lowe. 2002. Economic analysis and discussion of automatic sorting technology. Monograph. Teutopolis, IL: Farmweld, Inc. Gonyou, H. 2005, Winter. Handling the grow/finish process. Western Hog Journal (Special Interest). Online: www.albertapork.com.

Kelley, T. 2004, August 31. Auto sorts get a retrofit. Pork magazine.

Li, N., A. P. Schinckel, P. V. Preckel, and B. T. Richert. 2003. Using a stochastic model to evaluate swine production management with Paylean IV: Return of accurate sorting for marketing. Swine Research Report. West Lafayette, IN: Purdue University.

Hastad, C., M. Tokach,, S. Dritz, R. Goodband, J. DeRouchey, and J. Nelssen. 2005, Dec. 15. Sorting by weight, bumping energy improves returns. National Hog Farmer.

Vansickle, J. 2004, Nov. 15. Ten keys to automatic sorting. National Hog Farmer.

Wellock, I. J., G. C. Emmans, and I. Kyriazakis. 2003. Predicting the consequences of social stressors on pig food intake and performance. J. Anim Sci 81:2995-3007.

Wellock I. J., G. C. Emmans, and I. Kyriazakis. 2004. Modeling the effects of stressors on the performance of populations of pigs. J. Anim Sci 82:2442-2450.

April 2008