cristina giosue

Slide1: 

UNIVERSITY OF PALERMO - SICILY (Italy) Department S.En.Fi.Mi.Zo - Animal production Sector -   ANSC 400, CORNELL UNIVERSITY, 05/04/2006 Dr. Cristina Giosuè A System Dynamics application: “How to increase the total donkey milk production in Italy, producing profitability to the farmers?” PhD student in “PRODUZIONI FORAGGERE MEDITERRANEE” XVIII cycle Visiting fellow at Cornell University in the Departmnent of Animal Science

Outline: 

Outline Introduction to System Dynamics modeling Examples of System Dynamics application on a donkey milk production model

Slide3: 

Introduction to System Dynamics modeling

What is System Dynamics and why it is important?: 

What is System Dynamics and why it is important? - The complexity of the systems in which we live is growing (accelerating economic, technological, social, and environmental changes) - Many of the problems we now face arise as unanticipated side effect of our own past actions - All too often the policies we implement to solve important problem fail, make the problem worse, or create new problems CSDNet

Slide5: 

System Dynamics is - perspective and set of conceptual tools that enable us to understand the structure and dynamics of complex systems - qualitative modeling method that enables us to build formal computer simulations of complex systems and use them to design more effective policies and organizations - a rigorous way to help thinking, visualizing, sharing, and communication of the future evolution of complex organizations and issues over time for the purpose of solving problems and creating more robust designs, reducing the likelihood of counterintuitive or unintended consequences What is System Dynamics and why it is important? CSDNet

Slide6: 

Figure 1-3 in Sterman Assumes a clear and linear response of results to our decision Why unintended consequences? Event-oriented world view Every event has a (single) cause Leads to event-oriented problem solving

Slide8: 

System Dynamics involve the use of: Diagrams, graphs, words, and basic algebra to activate and capture existing knowledge about a particular observed problem situation Frameworks to help both researchers and practitioners organize, filter and structure that knowledge Mathematical simulation models and learning environments that help researchers and decision makers to identify more sustainable solutions and further refine their conceptual models CSDNet

Slide9: 

“All models are wrong, some models are useful” All models are simplifications and must omit elements of reality Despite these omissions, “good models” can enhance our thinking and problem solving All models have “clients” Clients are people you must influence for your work to have impact Help clients solve their problem……but also challenge their thinking

Slide10: 

Examples of System Dynamics application on a donkey milk production model

The SD Modeling Process: 

Iteration can occur from any step to any other Modeling effort will go through each of these step many times The SD Modeling Process

The SD Modeling Process: 

Iteration can occur from any step to any other Modeling effort will go through each of these step many times The SD Modeling Process Real world Information feedback Mental models of real world Strategy, structure, decision rules Decisions (Organizational Experiments)

Slide13: 

The researchable PROBLEM is: How to increase the total donkey milk production in Italy, producing profitability to the farmers?”…….. Problem articulation

Slide14: 

…..and WHY? Problem articulation

Slide15: 

General problem (background) Food Allergies a) In the last 20 years the population of developed countries suffering food allergies has increased (from 5 to 10%), the children and the babies are the most affected (www.italiasalute.it) b) FA develop when intestinal immune system doesn’t respond normally to the food proteins or protein fragments which have escaped lumen hydrolysis Problem articulation

Slide16: 

General problem (background) Food Allergies The main foods, that frequently cause allergy, are: egg (albumin), peanut, walnut,hazelnut, fish,shellfish, cow milk, chocolate, fruit (strawberry, pineapple, orange, tropical fruit ecc), cheese and vegetable (tomate ecc) ecc. The common symptoms of FA are: nausea, vomiting, abdominal pain, distension, flatulence and diarrhea; sometime skin and respiratory tract may be involved. Occasionally, severe systemic (anaphylactic) reactions are provoked and these reactions may be fatal (Sampson et al., 1996; Kimber et al., 2002). Problem articulation

Slide17: 

General problem (background) Cow Milk Protein Allergy (CMPA) Cow milk protein allergy (CMPA) is the most relevant with higher social implications in developed countries. Prevalence: 2-5% of the children and babies in the developed countries (Pizzin et al., 2003; Villoslada et al., 2005) and 1% in the adults (E. Smith, 1997). Problem articulation

Slide18: 

CMPA Most children with CMPA synthesize specific immunoglobulin E to proteinic antigens (Iacono et al., 1992), such as α, β and k caseins, β lactoglobuline, α lactoalbumine and lactoferine (Teschemacher et al., 1997; F. Lara-Villoslada et al., 2005; CMPA in infancy and childhood) General problem (background) Problem articulation

Slide19: 

CMPA Actual alternative feeding systems........ -Dietary products for infants derived from several different protein sources (such as bovine casein, bovine whey, bovine or porcine collagen, soy, or mixtures of these) exposed to different procedures of hydrolysis and further processing (heat treatment or ultrafiltration) -Dietary products based on amino acid mixtures (Fiocchi et al., 2003). General problem (background) Problem articulation

Slide20: 

CMPA …..but Some patients with CMAP can also react to these foods (multiple food allergies) (Iannolino et al., 2005) General problem (background) Problem articulation

Slide21: 

Researchable problem The donkey milk and CMPA Since 1990, in Italy, some researches in regard to donkey milk use on children with strong food allergies have started. In some regions of Italy the use of donkey milk to feed the babies is not new (in Sicily until the end of the Second War and in Germany) (Oftedal et al., 1988). The economical, social changes and the industrial progress have caused an increase in the use of milk substitutes. Iacono et al. (1992) showed good results by the use of donkey milk on 9 babies with multiple food allergies. The donkey milk was well tolerated and no negative reactions were recorded. These results were confirmed by Carroccio et al. (2000) on 18/21 patients. Problem articulation

Slide22: 

Milk composition (%) of different species and Energy (KJ/kg) (Polidori 1994) *Salimei et al., 1999; **Salimei et al, 2001; ***Polidori 1994

Slide23: 

Average nitrogen composition of the milk in some species

Slide24: 

Some Milk Proteic Fractions + Jenness, 1980; *Doreau et Boulot, 1989; **Buscemi, 2000;***Ambruzzi, 2003 More investigations are necessary!

Slide25: 

Acidic composition of the donkey milk fat (Chiofalo 2001)

Slide26: 

Essential Fatty Acids (EFAs) and human health EFAs are long-chain polyunsaturated fatty acids derived from linolenic (Omega-3), linoleic (Omega-6), and oleic acids (Omega-9) (The number following "Omega-" represents the position of the first double bond, counting from the terminal methyl group on the molecule) Omega-9 is "non-essential" because the body can manufacture a modest amount on its own, provided essential EFAs are present EFAs are necessary fats that humans cannot synthesize, and must be obtained through diet and support the cardiovascular, reproductive, immune, and nervous systems by the production of prostaglandines (they are fundamental for growth in fetuses (by mother’s dietary intake) and children, particularly for neural development and maturation of sensory system (www.goodfats.pamrotella.com)

Slide27: 

Essential Fatty Acids (EFAs) and human health

Conjugated linoleic acids (CLA) and human health: 

Conjugated linoleic acids (CLA) and human health Mixture of positional and geometric isomers of linoleic acids with two double bonds, which can be different in the position and orientation (cis or trans) located on adjacent carbons. The main source of CLA in human diets: -food products derived from ruminants (meat and dairy products) The CLA are showing impressive range of beneficial health effects in biomedical studies with animal models, like: Anti-carcinogenic Anti-obesity and altered nutrient partitioning Anti-atherogenic and reduces cholesterol Enhance the immune system Enhance bone mineralization (Bauman & Lock, 2006)

Conjugated linoleic acids (CLA): 

Conjugated linoleic acids (CLA) cis-9, trans-11 CLA and trans-10, cis-12 CLA cis-9, trans-11 conjugated linoleic acid is the predominant CLA isomer found in ruminant fat including milk fat. CLA is produced as an intermediate in rumen biohydrogenation of linoleic acid present in the cows diet and a portion of the CLA in milk fat comes from that which has escaped complete biohydrogenation in the rumen. However, the majority is made by the cow herself using the desaturase enzyme and trans-11 18:1 (vaccenic acid), another fatty acid intermediate produced during rumen biohydrogenation. With certain diets the rumen environment is changed and a portion of the biohydrogenation produces, trans-10, cis-12 CLA and trans-10 C18:1 as intermediates. trans-10, cis-12 CLA is present at only trace levels in milk fat (Bauman & Lock, 2006) cis-9, trans-11 CLA

Trans vaccenic and CLA content in milk fat of different species in comparison with human milk fat : 

Trans vaccenic and CLA content in milk fat of different species in comparison with human milk fat (Jahreis et al., 1999)

Collaborative Research project: effect of different diet and oxytocin use on donkey milk fat yield and fatty acids composition: 

Collaborative Research project: effect of different diet and oxytocin use on donkey milk fat yield and fatty acids composition INVESTIGATORS: Cornell University Department of Animal Science (USA) , Istituto Sperimentale Zootecnico per la Sicilia and University of Palermo Department SEn. Fi.Mi.Zo. (Italy) OBJECTIVES: To investigate changes in the pattern of donkey milk fatty acids, using two different diets, one of them with an extra virgin olive oil integration. To investigate changes in donkey milk fat yield and fatty acids composition, using oxytocin

Slide32: 

Researchable problem The donkey as livestock animal MILK Medicine: - PEDIATRICS - GERIATRICS - CARDIOLOGICS - FOR SOME TUMORS Cosmetic industry Problem articulation

Slide33: 

Researchable problem The donkey as livestock animal MEAT - brasato Stracotto - Smoked meat Cold meat and salami Problem articulation

Slide34: 

Researchable problem The EU is showing a particular attention to promote the livestock production in marginal areas, to prevent and to stop the abandon of these rural areas, which can cause very relevant and negative consequences on the environment, the economical and social aspect. The donkey is a breed in extinction in developed countries, and this animal is presented in marginal areas, thanks its high resistance and adaptation capacity Problem articulation

Slide35: 

Donkey breeds in Italy 6 main donkey breeds Martina Franca Sardo

Slide36: 

Donkey breeds in Italy 6 main donkey breeds Amiata Asinara Romagnolo

Slide37: 

Donkey breeds in Italy 6 main donkey breeds Ragusano Pantesco Sicilian

Slide38: 

The donkey, which has used in the past to work the land and to transport, now as a livestock animal can be an alternative profitability resource for the marginal areas of the Mediterranean as well as for many agriculture areas of the developing countries, In relationship with the other income by the sale of donkey for meat and by other possible utilizations (trekking, brain gym, pet therapy)

Slide39: 

Problem of Donkey milk sale in Italy The donkey milk is not included in the DPR 54/97 The sale can be only direct (from the farmer to the consumer) through autorization from the ASL (local sanitary agency) (Regio Decreto 9th May 1929 n° 994) The VETERINARY INSPECTORATE has institued in Sicily a TECHNICAL AND SCIENTIFIC COMMITTEE to investigate more donkey milk production and health aspects, to establish policies and marketing strategies improving this new productive sector

Slide40: 

Dynamic problem definition Define a “Reference Mode” Graphs or other information showing the development of the relevant problem over time The information about the donkey milk production and donkey milk consume is not registered by the statistical institutions in Italy; The donkey indeed is not included into animals producing milk but only meat, considering the low incidence of this product and the actual selling and production restrictions. Problem articulation

In Italy, since 1960, donkey number has decreased highly (FAO 2005). : 

In Italy, since 1960, donkey number has decreased highly (FAO 2005).

Qualitative reference mode of donkey milk production (liter/month) : 

Qualitative reference mode of donkey milk production (liter/month)

Qualitative future reference mode of donkey milk production (liter/month) : 

Qualitative future reference mode of donkey milk production (liter/month)

It is important to define the relevant time horizonenough of past to show problem developmentenough into future to show possible delayed effects of potential policies: 

It is important to define the relevant time horizon enough of past to show problem development enough into future to show possible delayed effects of potential policies In this case I have chosen 8 years like time horizon, because I think that this can be enough to evaluate the results of application of different policies and strategies, increasing the milk production and then the milk available for the market Problem articulation Dynamic problem definition

Dynamic Hypothesis: 

Dynamic Hypothesis A proposed explanation for the behavior Explicit structure that creates behavior Provisional, subject to revision (working theory) Endogenous focus Behavior arising from within the system Not (just) external shocks Avoid “narrow” model boundaries

Tools for Dynamic Hypotheses: 

Tools for Dynamic Hypotheses Model Boundary Chart What variables included, excluded What variables exogenous, endogenous Causal Loop Diagrams (CLD) Show causal linkages among variables Focus on feedback structure Stock-Flow Diagrams (SFD) Characterize physical stock-flow structure Dynamic Hypothesis

Population and birth rate loop: 

Population and birth rate loop INCREASE in population increases births, INCREASE in births increases population. This is a POSITIVE feedback loop, which will cause population to grow. Dynamic Hypothesis

Population and death rate loop: 

Population and death rate loop Population INCREASES death rate, death rate DECREASES population This is a NEGATIVE or BALANCING feedback loop Dynamic Hypothesis

Slide49: 

The simple system has two feedback loops These operate together to produce the behavior of the system. Population increases birth rate, birth rate increases population Population increases death rate, death rate decreases population Dynamic Hypothesis

Slide50: 

Dynamic Hypothesis

Slide51: 

Dynamic Hypothesis

Slide52: 

Dynamic Hypothesis

Slide53: 

Dynamic Hypothesis

Slide54: 

Dynamic Hypothesis

Slide55: 

Dynamic Hypothesis

Slide56: 

Dynamic Hypothesis

Slide64: 

Explicit delineation of stocks and flows Stocks (states) are accumulations Material or information Can be counted at a given time Change only through flows Flows (rates) are quantities per some amount of time Change states Cannot be measured instantaneously Can be affected by many other variables Dynamic Hypothesis

Slide65: 

“State-rate” structure Rates affect states, states affect rates Units consistency for variables, equations Explicit specification of units for each element Dynamic Hypothesis

Graphical Representation: 

Graphical Representation Dynamic Hypothesis

Graphical Representation: 

Graphical Representation FBR and AL are Auxiliary Variables (neither stocks nor flows) Population also determines the birth and death rates in this case Dynamic Hypothesis

Slide68: 

Stock and flows diagrams AGING CHAIN DONKEYS

Slide69: 

Stock and flows diagrams MILK PRODUCTION, MILK DEMAND AND MILK PRICE

Slide70: 

Stock and flows diagrams CONSUMER

Stock and flows diagrams FEED RESOURCE: 

Stock and flows diagrams FEED RESOURCE

Stock and flows diagrams DONKEY FARM SYSTEM: 

Stock and flows diagrams DONKEY FARM SYSTEM

Slide73: 

Stock and flows diagrams REVENUE, COSTS AND NET MARGIN

Mathematics of SD models: 

Mathematics of SD models System of ordinary differential equations Solved by numerical integration St = ∫(Inflow-Outflow) ds + S0 Inflow = f(S, other variables) Outflow = f(S, other variables) Many software programs available Vensim® is good for research purposes Dynamic Hypothesis

Testing (Formulate Simulation Model) : 

Testing (Formulate Simulation Model) An explicit mathematical representation helps to: Identify vague concepts Resolve contradictions previously unnoticed Provides a real test of understanding of the problem and its elements Specification must be complete and consistent Many softwares available I’m using Vensim® from Ventana Systems

Model Evaluation: 

Model Evaluation Begins with first equation Comparison of behavior in model to real world Concepts in model should correspond to a meaningful real world concept Check for Dimensional consistency Sensitivity to parameter changes Response to “extreme conditions”

Policy Design and Evaluation: 

Policy Design and Evaluation Modifications to relevant parameters Creation of entirely new strategies and structures Modifying the feedback structure Reducing or eliminating the delays Changing information flows Altering decision rules Assess sensitivity of policy results

And now a little bit of practice!: 

And now a little bit of practice! Milk production model

Slide79: 

Thank you !