Information Management for Pig Professionals

Jerry Shurson

Department of Animal Science, 335d Animal Science/Veterinary Medicine Bldg., 1988 Fitch Ave., University of Minnesota, St. Paul, MN, USA 55108

Introduction

Our biggest challenge to being effective and successful professionals is to develop an efficient and effective information management system. The amount of new information generated each day is growing exponentially. This, coupled with large reservoirs of underutilized, previously generated information are tremendous assets in our day-to-day decision making and advising. However, information by itself is the "raw" form of knowledge. As pig professionals, a main part of our daily activities involves acquiring, analyzing, interpreting and applying knowledge in a form that allows effective decision making. The information management tools available to us today provide rapid, efficient, and increased access to valuable information.

Evolution of Information Management

Information Sources

Information provided on paper continues to be the predominant form for exchanging and acquiring new information. However, it is also the least efficient. Libraries are filled with rows and rows of paper text, some of which is accessed very infrequently. Although libraries have served as an invaluable "information warehouse," access to relevant information, timeliness, convenience, and storage space, are limits to our information management needs.

With the rapid development and adoption of electronic information management, assembling related information on CD ROM has become a much more efficient way to store and retrieve information. However, CD ROM only functions as a more efficient and convenient way of organizing, storing, and retrieving "archived" information that already exists in some form. Thus, even though information management is dramatically improved through the use of CD ROM technology, the value of the information is limited to the time frame in which it was assembled.

Access to the rapidly growing amount of information on the World Wide Web (www) is increasing at a phenomenal rate. The www provides many of the same information management features of CD ROM (i.e., convenience, organized information search, retrieval speed, and access to "information archives"), but also provides access to more current, real-time information on every conceivable topic of interest. The primary disadvantages of using the www as the sole information management tool are the lack of access to all relevant information sources, and limited quality control of the information available. Although there are a relatively large number of agriculturally related web sites, sites containing information specific to pigs are limited. Furthermore, you need to know where to look to find these sites. Search engines provide this capability, but the large amount of time required to travel through potential sites in search of information is a major limitation for obtaining information when you need it.

Value Added Information Sources

Most information by itself is of limited value unless it is evaluated for quality and relevance, converted into knowledge and applied to "real life" situations. Each of these processes adds value to the information available. For the most part, we, as pig professionals, are responsible for performing these functions. Those who are capable of being innovative and performing these value added functions will lead technology adoption in the pork industry.

Computer Software. Computer software developers have applied these processes with varying degrees of success to specific information management/information integration/decision support functions. However, the most widely used computer software packages are production and financial accounting packages which do little to add value to raw data. As a result, we developed SwinePro to demonstrate how minimum economic and performance data could be converted into knowledge for making profit improving management decisions unique to a specific pork production operation. To do this, we designed the software to require economic and production data input, compare existing performance to benchmarks, prioritize the most economically significant opportunities for improvement, identify solutions to performance problems, and allow the user to create "what if" scenarios with their own farm specific data set.

One of the fundamental problems we have with farm based record information is lack of uniformity of definitions and formulas (production and financial standards) which makes constructing, comparing and interpreting databases a challenging, often misleading, and cumbersome task. The NPPC Production and Financial Standards will greatly improve our ability to compare and evaluate databases if the standards are widely adopted by software companies and the entire industry. This is a critical step in the continual process of attempting to develop and use software to convert data into knowledge for application.

Models. Most recently, mathematical models used for determining nutrient requirements have become one of the most widely discussed electronic information management tools, despite the fact that models have existed for many years. The primary reason for this is that most earlier models were developed by academic professionals for research purposes, and their application to "real world situations" was poor due to lack of enough information to deliver the expected amount of precision in the derived estimates. Another reason for the poor adoption and imprecision has been the detail of scale used to develop the model (cell vs. tissue vs. organ vs. body vs. population on-farm vs. country population). Newer models have generally focussed on data that can be measured on a farm to provide estimates of gross inputs needed to achieve a desired response without accounting for and describing all of the interactive elements of the system. The problem with this approach is that we are usually able to develop reasonable estimates of nutrient requirements to maximize performance, but have no assurance that the diet formulas being fed maximize net profit. One exciting potential application of using models for information management and decision support, involves supply optimization modelling technology. Several non-agricultural firms have developed, and are using, sophisticated supply optimization models to define the necessary inputs required to achieved a desired output (rate of return on investment). This approach appears to have tremendously powerful potential in defining the necessary genetics, diets, facilities and environment to achieve a specific end product at an acceptable rate of return on investment. In this context, diets would be formulated to optimize, not maximize, performance.

Expert Systems. We have experimented with the application of expert systems technology to integrate and identify solutions to pig production problems based on a database representing existing performance and financial performance in a swine operation by developing SwinePro (Klee et al., 1994). In theory, one should be able to construct a "decision tree" with various possible outcomes to arrive at a final solution to a problem. However, with all of the interactive elements in a pig production system, and the difficulty of adequately describing them and weighing their importance, construction of a comprehensive, precise expert system to produce farm specific recommendations to improve profitability is difficult.

Neural Networks. Neural networks are another exciting information management tool that has application in identifying patterns in databases and learning about the interactive relationships among variables in complex systems. We have experimented with these tools using PigCHAMP and Nebraska Swine Enterprise Record databases to evaluate their utility for information management in decision support (Leland and Shurson, 1993). We learned two important things when we applied neural networks to these two common databases: 1) our current inability to describe multidimensional components of pig production systems with sufficient detail limits the value of this tool, and 2) due to the linearity of our production and financial datasets, use of more commonly used statistical tools such as multiple linear regression models are more appropriate.

Listservers. Another value added, information management tool involves the development and management of swine specific list servers. An example of one of these is Swine-L. These list servers are ways of electronically networking pig professionals to exchange questions, answers and experiences on a wide variety of swine topics. This has proven to be an excellent way of obtaining quality information that applies to an application question. Information is good because it is "peer reviewed" by all users as it is being shared. Any errors in recommendations are usually quickly pointed out by experts linked to the listserver. The major disadvantages of this information management method are that there is not an immediate response to a question, participants must be committed toward devoting time to respond or "chat," and many listservers have limited access (i.e., only selected individuals are allowed to subscribe and participate).

Information Needs of Pork Producers

One of the most important components of developing and revising your information management system involves understanding the needs of pork producers. Recent surveys have been conducted in the U.S. by the National Pork Producers Council, and specifically in Minnesota by the Minnesota Pork Producers Association, to determine information needs and priorities.

Relative rankings of areas of most critical importance to Minnesota Pork Producers were:

1. Management of the production system

2. Buildings and environment

3. Disease and health

4. Farm and business management

5. Nutrition

6. Marketing

7. Pork industry

8. Labour management

Notice that the area of greatest need for information (production systems management) involves integrating information from various disciplines to improve the efficiency of the system. For most pig professionals, our fundamental training has been in a traditional discipline (e.g., nutrition, health, environment, etc.) which has focussed our expertise in a specific area; but pork producers need assistance with putting all of the component parts of the system together to improve their efficiency and competitiveness. This is difficult for any individual to do. Thus, we need to network with other professionals to form advisory/consulting teams to improve the quality of information provided to producers.

Some other interesting statistics related to the current and planned use (in five years) of electronic communication and information technology of Minnesota pork producers are as follows: today 2002

Use a computer in pork business 63% 85%

Access to Internet/www on computer 25% 73%

Send and receive e-mail 26% 71%

Subscribe to an electronic news service 36% 61%

Almost two-thirds of Minnesota pork producers use a computer in their pork production business, and even a higher percentage plan to use computers in the next five years. Although this finding may not be surprising, it does indicate that the majority of pork producers recognize the benefits of using computers as part of their information system. Those who don't will be at a significant competitive disadvantage for acquiring and applying new knowledge to their production system. These statistics also indicate that pork producers plan to use electronic technology as part of their information management portfolio. As pig professionals, we also have the responsibility of utilizing these tools and providing information resources in these forms to pork producers.

Is There a Perfect Information Management System?

As you might expect, in the area of information management, there is no perfect system. However, the single, most important determinant of success in the pork industry, now and in the future, will be access to and use of knowledge (DiPietre, 1996). Therefore, it is critical that we develop a personalized information management system that allows us to identify and acquire high quality information that is relevant and timely. Part of this process involves being aware of the multiple ways of acquiring and using information when we need it. Because of the speed, efficiency and convenience of electronic technology, we need to be using more of these tools as part of our information management portfolio. The complexity, sophistication, and specialization of the pig industry require us to determine the limits of our expertise and determine ways of constructing a personalized information management system that meets the information needs of our daily job responsibilities. To do this, I suggest a simple five-step process to develop a more formalized and effective information management system.

1. Inventory your current information sources.

2. Inventory your current information needs.

3. Inventory and rate your core competencies.

4. Compare steps 1, 2, and 3 to determine deficiencies.

5. Identify traditional and electronic information sources that will improve your ability to find relevant information when you need it.

References

DiPietre, D. (1996) Current mega trends in the U.S. pork industry. In: NPPC Swine Educators Conference Proceedings, pp. 1-10.

Klee, A., J. Shurson and J. Lemon. (1994) SwinePro Documentation. © University of Minnesota and Ohio State University.

Leland, P., and J. Shurson. (1993) Neural network applications in swine production. Recent Advances in Swine Production and Health 3:72.